UNIVERSITI TEKNOLOGI MALAYSIA

NOTES : * If the thesis is CONFIDENTAL or RESTRICTED, please attach with the

letter from the organization with period and reasons for confidentiality

or restriction.

DECLARATION OF THESIS / UNDERGRADUATE PROJECT PAPER AND COPYRIGHT

Author’s full name : LEE CHENGXU

Date of birth : 16 MARCH 1988

Title : DESIGN OF PLASMA NEEDLE AND HIGH FREQUENCY

POWER SUPPLY FOR BIO-MEDICAL APPLICATIONS

Academic Session : 2011/2012

I declare that this thesis is classified as :

CONFIDENTIAL (Contains confidential information under the

Official Secret Act 1972)*

RESTRICTED (Contains restricted information as specified by

the organization where research was done)*

OPEN ACCESS I agree that my thesis to be published as online

open access (full text)

I acknowledged that Universiti Teknologi Malaysia reserves the right as follows :

1. The thesis is the property of Universiti Teknologi Malaysia.

2. The Library of Universiti Teknologi Malaysia has the right to make copies for

the purpose of research only.

3. The Library has the right to make copies of the thesis for academic

exchange.

Certified by:

SIGNATURE SIGNATURE OF SUPERVISOR

880316-02-5377

ASSOC. PROF. DR. ZOLKAFLE BUNTAT

(NEW IC NO. /PASSPORT NO.) NAME OF SUPERVISOR

Date : 27th JUNE 2012 Date : 27th JUNE 2012

√

“I hereby declared that I have read this thesis and in my

opinion this thesis is sufficient in terms of scope and quality for the

award of Bachelor of Engineering (Electrical).”

Signature :

Name of Supervisor : ASSOC. PROF. DR. ZOLKAFLE BUNTAT

Date : 27th

JUNE 2012

DESIGN OF PLASMA NEEDLE AND HIGH FREQUENCY POWER

SUPPLY FOR BIO-MEDICAL APPLICATIONS

LEE CHENGXU

This report is submitted in partial fulfillment of the

requirements for the award of the degree of

Bachelor of Engineering (Electrical)

Faculty of Electrical Engineering

Universiti Teknologi Malaysia

JUNE 2012

ii

I declare that this thesis entitled “Design of Plasma Needle and High Frequency

Power Supply for Bio-Medical Applications” is the result of my own research except

as cited in the references. The thesis has not been accepted for any degree and is not

concurrently submitted in candidature of any other degree.

Signature :

Name : LEE CHENGXU

Date : 27th

JUNE 2012

iii

Dedicated to my beloved father and mother, siblings, friends and lecturers for their

endless loves, encouragement and support

iv

ACKNOWLEDGEMENT

First of all, I would like to express my heartiest gratitude to my supervisor,

Assoc. Prof. Dr. Zolkafle Buntat for his professional guidance, comments and

endless inspirations in the preparation of this research from scratch to successfully

accomplish. I appreciated that I can still seek for his help, advices and suggestions

while in the middle of his busyness. Besides, I would like to thank Dr. Muhammad

Abu Bakar Sidik for his guidance and help as well.

My deepest appreciation goes to my family, friends, fellow course mates and

laboratory technicians for their endless love, endurance and support during the entire

research process. Last but not least, I would like to thank those who directly and

indirectly contribute in making this research a success.

v

ABSTRACT

Non-thermal plasma treatment of living tissues has become a popular topic in

medical sciences after it was discovered to have the bacteria inactivation function in

year 2002. It is generated under atmospheric pressure and room temperature by

radio-frequent excitation. At the moment, commercialized research function

generator and amplifier are used to generate the non-thermal plasma. Hence, the cost

for current high frequency power supply is very expensive and unfeasible to be used

widely in bio-medical applications. Hence, a cheaper high frequency power supply

should be developed for future applications. Besides, the generated plasma at the tip

of the current plasma needle also very small. It needs a longer treatment time if it is

being applied on a larger wound. In this work, a high frequency power supply for

non-thermal plasma source was designed and developed by using modified class-E

power amplifier circuit. MAX038 was used as signal generator to generate the high

frequency signal. MOSFET Driver DEIC515 and MOSFET DE275X2-102N06A

were used for ultra-fast switching before attached to the amplifier circuit. Based on

the simulation, the designed high frequency power supply has the capability of

generating 505Vpp at frequency of 13.56MHz and this output is high enough to

generate the non-thermal plasma. A novel design of plasma needle with a ring

magnet at the head of the needle was designed with the aims to improve the plasma

uniformity as well as increasing the healing effects on the treated area. Besides high

frequency power supply, this report will present the enhancement design of plasma

needle as well. Some recommendations for future research also being included in the

final part of the report.

vi

ABSTRAK

Baru-baru ini, rawatan plasma sejuk pada tisu hidup telah menjadi satu topic

yang popular dalam bidang sains perubatan selepas ia ditemui mempunyai fungsi

penghapusan bakteria pada tahun 2002. Ia dihasilkan di bawah tekanan atmosfera

dan suhu bilik oleh getaran frekuensi radio. Pada masa ini, penjana dan penguat yang

dikomersialkan untuk tujuan penyelidikan telah digunakan untuk menjana plasma

sejuk. Oleh itu, penjana kuasa berfrekuensi tinggi yang lebih murah perlu direkacipta

untuk aplikasi pada masa depan.Selain itu, saiz plasma yang dijana pada hujung

jarum plasma juga sangat kecil. Ia memerlukan masa rawatan yang lebih panjang jika

digunakan pada luka yang saiznya lebih besar. Dalam karya ini, penjana kuasa yang

berfrekuensi tinggi telah direka dan dibangunkan dengan menggunakan litar penguat

kuasa kelas E yang telah diubahsuai. MAX038 telah digunakan sebagai penjana

isyarat untuk menjana isyarat berfrekuensi tinggi. Pemandu MOSFET DEIC515 dan

MOSFET DE275X2,-102N06A telah digunakan untuk tujuan pensuisan ultra-cepat

sebelum dipasang pada litar penguat. Berdasarkan simulasi yang telah dibuat, rekaan

penjana kuasa frekuensi tinggi ini mampu menjana 505Vpp pada frekuensi

13.56MHz dan keluaran ini adalah cukup tinggi untuk menjana plasma sejuk. Reka

bentuk baru jarum plasma dengan magnet cincin di kepala jarum telah direka

bertujuan untuk meningkatkan keseragaman plasma serta meningkatkan kesan

penyembuhan pada kawasan yang dirawat. Selain penjana kuasa frekuensi tinggi,

laporan ini juga akan membentangkan jarum plasma yang direka. Beberapa cadangan

untuk kajian pada masa hadapan juga dimasukkan dalam bahagian akhir laporan ini.

vii

TABLE OF CONTENTS

CHAPTER TITLE PAGE

DECLARATION

ii

DEDICATION iii

ACKNOWLEDGEMENTS iv

ABSTRACT v

ABSTRAK vi

TABLE OF CONTENTS vii

LIST OF FIGURES x

LIST OF SYMBOLS xii

LIST OF ABBREVIATIONS xiii

LIST OF APPENDICES xiv

1 INTRODUCTION

1.1 Introduction 1

1.2 Background 1

1.3 Problem Statement 3

1.4 Objectives 4

1.5 Scope of the Project 5

1.6 Significance of the Study 5

2 LITERATURE REVIEW

2.1 Introduction 6

2.2 Plasma Generation 6

2.3 Plasma Needle 7

viii

2.4 Design of Plasma Needle 10

2.5 Class E Amplifier as RF Plasma Source 11

2.6 45MHz MOSFET Driver 14

2.7 MOSFET 17

2.7.1 MOSFET Turn-on Phenomena 18

2.7.2 MOSFET Turn-off Phenomena 21

2.8 Magnetic Effects on Living Organism 22

2.9 Application of Plasma Needle 23

2.9.1 Dental Applications 23

2.9.2 Plasma Treatment on Mammalian Vascular

Cells

24

2.9.3 Cancer Treatment 24

3 RESEARCH METHODOLOGY

3.1 Introduction 25

3.2 Methodology Procedure 26

3.3 Related Guidelines and Datasheets 27

3.4 Software Used for Modelling 27

3.4.1 SolidWorks 2011 27

3.4.2 Multisim 10.0 29

4 PLASMA NEEDLE DESIGN

4.1 Introduction 30

4.2 Modelling Components 30

4.3 Modelling Dimensions 31

4.4 Modelling Descriptions 32

5 HIGH FREQUENCY POWER SUPPLY

5.1 Introduction 35

5.2 Simulation Development 35

ix

5.2.1 Simulation of Modified Class E Amplifier 36

5.3 Hardware Development 37

5.3.1 MAX038 Frequency Waveform Generator 38

5.3.2 DEIC515 MOSFET Driver Circuit 39

5.3.3 DEIC515 and DE275X2-102N06A MOSFET

Circuit

40

6 RESULTS AND DISCUSSIONS

6.1 Introduction 42

6.2 Simulation Results for Modified Class E Amplifier 42

6.3 Hardware Results 44

6.3.1 MAX038 Frequency Waveform Generator 44

6.3.2 DEIC515 MOSFET Driver 45

6.3.3 DEIC515 and DE275X2-102N06A MOSFET 47

7 CONCLUSION AND RECOMMENDATION

7.1 Introduction 49

7.2 Conclusion 49

7.3 Recommendation 50

REFERENCES 52

APPENDICES

APPENDIX A 55

APPENDIX B 57

x

LIST OF FIGURES

FIGURE NO. TITLE PAGE

1.1 Thermal Plasma 2

2.1 General Schematic of RF Capacitively Coupled

Plasma

7

2.2 Schematic of Plasma Needle Setup 8

2.3 Stability Curves of The Plasma 9

2.4 Plasma Needle 10

2.5 Class E Amplifier Circuit Diagram 12

2.6 Experimental waveform 14

2.7 Symbol and equivalent circuit of a MOSFET 17

2.8 Transfer characteristics of a power MOSFET 18

2.9 A MOSFET being turned on by a driver in a clamped

inductive load

19

2.10 A MOSFET being turned off by a driver in a clamped

inductive load

19

2.11 MOSFET turn on sequence 20

2.12 MOSFET turn off sequence 22

4.1 Components of Plasma Needle 31

4.2 Dimensions of Plasma Needle 31

4.3 Plasma Needle Model 33

4.4 The Pyrex tube moved 5mm into the plastic holder 34

4.5 Plasma Needle after Rendered 35

5.1 Modified Class E Amplifier Simulation Circuit 36

5.2 MAX038 Frequency Waveform Generator Circuit 38

5.3 MAX038 Hardware Circuit 39

xi

5.4 DEIC515 MOSFET Driver Circuit 39

5.5 DEIC515 Hardware Circuit 40

5.6 DEIC515 and MOSFET Hardware Circuit 41

6.1 Simulated waveform of VCT and VCR 43

6.2 Simulated waveform of Vin and Vo 43

6.3 Output of MAX038 44

6.4 Output of DEIC515 46

6.5 Output of DE275X2-102N06A 48

xii

LIST OF SYMBOLS

K - Kelvin

0C - Degree Celcius

UV - Ultraviolet

Min - Minute

ns - Nanoseconds

IG - Gate Current

ID - Drain Current

IS - Source Current

V - Volts

kV - Kilo Volts

kΩ - Kilo Ohms

mm - Milimeter

cm - Centimetre

mV - Mili Volts

mW - Mili Watts

GHz - Giga Hertz

MHz - Mega Hertz

Vpp - Peak-to-peak Voltage

Vp - Peak Voltage

He - Helium

uH - Micro Henries

mH - Mili Henries

ZL - Load Impedance

VGS(th) - Threshold Voltage

VCC - Power Supply Voltage

xiii

LIST OF ABBREVIATIONS

RF - Radio Frequency

f - Frequency

BNC - Bayonet Neill–Concelman

DC - Direct Current

AC - Alternating Current

UTM - Universiti Teknologi Malaysia

CCP - Capacitvely Coupled Plasma

NI - National Instrument

OD - Outer Diameter

ID - Internal Diameter

IGBT - Insulated Gate Bipolar Transistor

MOSFET - Metal–oxide–semiconductor Field-effect Transistor

IC - Integrated Circuit

PCB - Printed Circuit Board

xiv

LIST OF APPENDICES

APPENDIX TITLE PAGE

A Apparatus for Hardware Experiment 55

B Datasheet 57

CHAPTER 1

INTRODUCTION

1.1 Introduction

The conducted research was Design of Plasma Needle and High Frequency

Power Supply for Bio-Medical Applications. The reason this topic was chosen is

because non-thermal plasma treatment has become more and more popular in

modern plasma physics and in medical sciences. Besides, non-thermal plasma has

made an innovative form in solid state processing technology. In this chapter, there

are a few things that will be covered, namely background of the study, problem

statement, objective of the study, scope of the study and significance of the study.

1.2 Background of the Study

On earth, solid, liquid and gas are three states of matter that can be easily

found. Human are very familiar with these three states of matter as they are facing

them everyday. In 1879, an English physicist, Sir William Crookes had identified a

2

fourth state of matter where we call it as plasma now. The name plasma was first

being applied by Dr. Irving Langmuir, an American chemist and physicist, in 1929.

Plasma comprises of free charge carriers (electrons and ions), active radicals

and excited molecules and atoms which is similar to gas. High energy such as

thermal, electrical, or light is needed to ionize its atom or molecules and this process

will cause the gas to become electrically conductive. This electrically conductive,

ionized gas is called plasma. The temperature for plasma electrons normally above

104 K while the temperature for neutrals and ions depend on type of plasma and have

the range from room temperature to 107 K.

Plasma can be classified as „thermal‟ and „non-thermal‟ based on their

relative temperatures of the electrons, molecules, ions and atoms. For thermal plasma,

its electrons temperature and gas temperature are in thermodynamic equilibrium. In

other word, the temperature of molecules, ions and atoms are same with the electrons

temperature. Few examples of thermal plasmas are shown in Figure 1.1.

(a) (b)

Figure 1.1: Thermal Plasma (a) Lighning Strike and (b) Metal Spraying

3

For non thermal plasma, the electron temperature and gas temperature are not

in equilibrium with each other. The temperature of molecules, ions and atoms are

much lower (normally room temperature) than electrons (higher temperature). The

relative low temperature of non-thermal plasma make it does not impose thermal

damage to nearby object. The thermal damage caused by the high temperature of

thermal plasma has been eliminated or minimised. Since the thermal damage had

been eliminated, it can be applied in many applications.

The current trend is more focus on using non-thermal plasma in bio-medical

applications, usage on living tissues and even usage on human body. Non-thermal

plasma has the capability of bacterial inactivation [1], non-inflammatory tissue

modification [2] and healing effects on a living organism [3][4]. These capabilities

led to the development of the plasma needle and many researchers glow their interest

at the interaction between non-thermal plasmas and biological tissues.

1.3 Problem Statement

The plasma system that researchers are using now is very expensive. This is

because commercialize function generator and research amplifier are used in plasma

radio-frequency (RF) power supply. This equipment is not specifically use for

generating non-thermal plasma but normally used in laboratory with the purpose of

studying and researching. Due to the high cost of the plasma system, alternative for

cheaper RF power supply is needed so that the cost for plasma system can be

lowered down. This is vital for future applications of non-thermal plasma.

One of the applications for non-thermal plasma is plasma needle which

introduced by Eva Stoffels in 2002 [5]. The generated plasma is non-thermal plasma

where the temperature of the molecules, ions and atoms is about room temperature

4

and operates at atmospheric pressure. These characteristics open the opportunity of

using plasma needle in biomedical fields and even on human body. However, plasma

that is generated by plasma needle setup is very small (about one milimeter in

diameter). Due to this reason, the treatment time will be prolonged when the

treatment area is enormous.

It is proven that non-thermal plasma has the healing effects on a living

organism [3]. Medical research states that placing the N-pole side (the magnetic

negative energy of the two poles) of a magnet facing any living tissue may attract the

positive electrolytes in that living organism. By placing the N-pole side of a magnet

on the skin, this process is accelerated, and endorses natural healing. The

introduction of the ring magnet with north pole facing the treated surface, at the head

of the plasma needle is expected to improve the plasma uniformity as well as

functioning as healing effect.

1.4 Objective of the Study

The aim of this project is to study, research and redevelop the high frequency

power supply and design a new plasma needle by adding a ring magnet to the plasma

needle to improve the plasma uniformity. This aim will be met through two

objectives:

i. To redevelop a high frequency power supply for plasma needle;

ii. To enhance the design of plasma needle by ring magnet addition.

5

1.5 Scope of the Study

The scope of this project is:

i. Develop the high frequency power supply for plasma needle;

ii. Simulation study and modelling on high frequency power supply circuit;

iii. Analysis on the power supply output;

iv. Enhance the design of plasma needle by ring magnet addition.

1.6 Significance of the Study

Although there are various experiments being carried out on the plasma

needle, they did not discuss on the ways of improving the plasma uniformity and

treatment area. This study is to enhance the design of plasma needle by adding a ring

magnet with the purpose of improving the plasma uniformity as well as the healing

effects on the treated are. By improving the plasma uniformity and treatment area,

the treatment time can be shortening and more patients can be benefitted every day.

The findings of this study were important as it opens the ways of improving plasma

uniformity for future research.

6

CHAPTER 2

LITERATURE REVIEW

2.1 Introduction

In this chapter, literature review starts with screening the non-thermal plasma

generation, design and applications of plasma needle, which is the big picture of the

research. This research is part of the effort to generate non-thermal plasma for the

usage of plasma needle.

2.2 Plasma Generation

Plasma can be generated from various ways. The most common method in

plasma generation is the electrical breakdown of an electric field in a neutral gas. In

the electric field, the speeded charge carriers will transfer their energy into the

plasma by smashing other atoms and molecules. The electrons will preserve their

major energy in elastic collisions with other particles due to their light weight and

they will only transfer their energy mainly in inelastic collisions. The plasma

generated by the electric fields can be classified into four which are direct current (dc)

7

discharges, pulsed dc discharges, radio-frequency discharges and microwave

discharges [6].

The RF discharges can be divided into inductively coupled discharge and

capacitively coupled discharge. Plasma needle is using the capacitively coupled

discharge method to generate plasma in its application. Figure 2.1 shows the general

schematic of an RF capacitively coupled discharge.

Figure 2.1: General Schematic of RF Capacitively Coupled Plasma [7]

2.3 Plasma Needle

In year 2002, Physicist Eva Stoffels and her team had come out with an

innovative idea which is the plasma needle. Plasma needle is a novel design of non-

thermal plasma source which being generated at atmospheric pressure by using the

concept of radio-frequency discharges. It has a single-electrode configuration and

operated with the presence of helium gas [5]. The plasma generated in plasma needle

operates near room temperature and at atmospheric pressure, do not cause pain and

bulk destruction of the tissue yet allows treatment of uneven surface and has a small

penetration depth. These features allow plasma needle to be applied in bio-medical

applications.

8

RF plasma at 13.56MHz is generated by using a waveform generator. The

output from waveform generator is amplified by an RF amplifier. The signal is then

directed to a matching network. The power is monitored by using a power meter and

Dual Directional Coupler while the voltage is measured via Tektronics probe. The

electrical measurements show that plasma needle operates at quite low voltages from

200-500 Vpp and the power dissipation range from 10 mW to at most a few watts.

Figure 2.2: Schematic of Plasma Needle Setup [7]

This plasma source will generate the plasma which contains free electrons

and ions, various chemical reactive species and energetic UV photons [8]. UV

emission and density of chemical reactive species are significant in defining the

performance of plasma in the treatment of biological materials. The spore

inactivation with an atmospheric pressure discharge can be mainly due to UV

radiation. With the absence of UV emission, action of chemical reactive species such

as O-, OH

-, N2

+, N2 and He can cause spore inactivation as well. Hence, the spore

inactivation depending on the operating conditions, it can be achieved either under

dominant UV radiation or under purely action of the reactive species [9].

The plasma generation with the presence of helium gas are most stable and

have the widest range of operating conditions. The operating conditions of the

9

plasma needle as a function of helium flow rate and percentage of air admixture at a

constant total flow rate are shown in Figure 2.3. This is mainly essential for saving

budgets and convenience of operation in small model openings. Since with the

presence of helium gas, the plasma has very low power dissipation and the sustaining

voltage is tolerable, it is preferable to be used in biomedical applications.

Figure 2.3 Stability Curves of The Plasma: (a) as a function of helium flow rate and

(b) as a function of percentage of air admixture at a constant total flow rate (350 ml

min-1

). Displayed are the breakdown voltages, needed to ignite the plasma (),

minimum operating voltages, just above extinction threshold () and maximum

voltages, just below arcing () [5].

10

2.4 Design of Plasma Needle

Plasma needle has relatively simple design. The tungsten needle of 0.3mm is

inserted coaxially in a Perspex tube which has an inner diameter of 4mm [10]. The

tungsten needle serves as the powered electrode here. The Perspex tube is put into

the stainless steel holder and it is protrudes from the stainless steel holder. The length

of the needle normally is around 6-8cm. The electrode is insulated by the Perspex

tube to prevent a discharge along the needle.

The breakdown voltage to generate plasma can easier be achieved with the

presence of gas helium at minimum peak-to-peak RF voltage of around 200V. The

RF signal voltage is connected to plasma needle through a coaxial cable. There is a

gas inlet in the stainless steel holder for gas helium to flow in. Helium is regulated by

a mass flow controller to flow at a rate of 21/min into the Perspex tube [10]. It will

mix with a small amount of air at the tip of the needle. The helium-air mixture is

significant to create active radical species for sterilization [4].

Figure 2.4: Plasma Needle

11

The needle should keep as short as possible. This is because the changes in

the power supplied to plasma needle will cause the changes in matching network.

The heating of the needle is thought to be the main reason of the changes. It is better

for coaxial cable that connecting the power supply to the plasma needle to cover

most of the distance rather than lengthen the needle itself. Even though the plasma

kills most of the bacteria, the plasma needle should be sterilized after the treatment

[18]. This is the precaution step taken to avoid any bacteria remain at the needle after

the treatment.

There is heat at the tip of the needle when plasma is formed. The damage will

be done when there is direct contact between the tip of the needle and the skin.

Besides, the tip is very sharp as well. Hence, contact with the skin should be avoided

[18]. The housing covers should cover the needle in total so that the metal tip will

never touch the skin. If there is accident contact made during the treatment, the cover

will touch the skin instead of the warm metal tip.

2.5 Class E Amplifier as RF Plasma Source

Class E amplifier is the improvements and adaptations from typical topology

with the purpose of generating high frequency ac signal. It has higher efficiency

which is around 85% compared to 70% of conventional class B or class C amplifier.

Naturally, class E amplifier has smaller power losses by a factor of 2.3 as compared

to conventional class B and class C amplifier with same transistor at same frequency

and output power [13].

12

Figure 2.5: Class E Amplifier Circuit Diagram

The way of generating excitation voltage was based on the modified class E

amplifier where it is different with the original topology. This configuration uses a

metal oxide/semiconductor field-effect transistor (MOSFET), which perform on-off

operation by driving a parallel resonant circuit. The output signal from square wave

generator (CGS3311) will be supplied to a driver, which provides the „on‟ and „off‟

pulses for the MOSFET [15]. The MOSFET output pulses will then be converted

into a sinusoidal high voltage signal by parallel RLC resonant (class E amplifier)

circuit.

In this class E amplifier circuit, the resistive component R or load charge Z is

connected in parallel to the resonant capacitor C. The main modification of this

amplifier compare to the classical amplifier is the resonant capacitor. In this

amplifier, the resonant capacitor works like an ac voltage supply with respect to the

load charge [14]. Consequently, the circuit amplifies the voltage signals that will be

applied to the load.

The amplifier discontinuous conduction mode is determined by the two

possible power switch operating modes. During the ON state (S = ON), the resonant

circuit is only governed by LR and CR, with CT playing the role of a voltage supply.

Thus, the frequency response is

√ (2.1)

13

During the OFF state (S = OFF), the resonant circuit is governed by LR, CR

and CT. The current signal I will supply the resonant circuit and the frequency

response is

√ [

] (2.2)

The transistor acts as a switch with a duty ratio D and a work frequency f is

limited by (f1 < f < f2). The S can be expressed by

S = (2.3)

The circuit will show two distinct frequencies according to the state of S. The

capacitive parameter for MOSFET will be fixed according to MOSFET manufacture.

Hence, CT has a fix value which can be found from the datasheet. Thus, the resonant

network parameter LR and CR can be obtained by

*

+ (2.4)

(2.5)

ON wt, for 0 < wt ≤ 2𝜋𝐷

OFF wt, for 2𝜋𝐷 < wt ≤ 2𝜋

14

(a)

(b)

Figure 2.6: Experimental waveform of (a) VCR(t), VCT(t), and (b) iLR(t) and IRS(t) [15]

2.6 45MHz MOSFET Driver

Out of many ways of driving MOSFET, IC Drivers offer convenience and

features that attracting many designers. The main advantage of IC Driver is

compactness where it has much smaller sized circuits. Besides, IC Drivers

15

intrinsically offer shortest propagation delay hence signals have smaller traverse

distance. It also has shorter rise and fall times. Since all important parameters are

specified in an IC Driver, designers manage to save their time and capital as they

need not go through process of defining, designing and testing circuits to drive the

MOSFET [19].

A proper planning and execution is needed in MOSFETS operation of Class

D or E amplifier and applications that require ultra-fast rise and fall times. The four

important elements are circuit loop inductance, power supply bypassing, circuit

layout, adequate grounding and shielding [19]. The inductive term will be created in

the loop of current path from power supply positive to ground. This loop must be

kept as short as possible by using few tiny capacitors and solder them neatly on the

Vcc and ground pins of Driver IC. Driver IC's bypass capacitor value can be

calculated by:

(2.6)

Where, = quiescent current drawn from Vcc

d = Duty cycle of the PWM waveform

Qg = Total gate charge of the MOSFET

Fsw = Switching frequency

Vripple = Tolerable ripple level on the Vcc

Driver circuit also needs a proper grounding. Loops should be avoided in

driver circuit. Drivers need a very low impedance path for current return to ground.

The three paths of returning current to ground are: 1. between driver IC and the logic

driving it; 2. between the driver IC and its own power supply; 3. between the driver

IC and the source emitter of MOSFET being driven [19]. All these paths should be

kept short in length to reduce inductance and be as wide as possible to reduce

16

resistance. A copper plane in a multi-layered PCB can be used to provide a ground

surface under the gate drive circuit. This ground plane should be connected to the

power ground plane of MOSFET source or emitter terminal to avoid different ground

potentials.

With desired rise and fall times in the range of 2 to 3 ns, lengths of current

carrying conductors should be kept as short as possible. Conductor trace‟s partial

inductance can be calculated by:

(

) (2.7)

Where, = inductance in nanohenries

h = height of conductor trace above ground plane

w = width of the conductor trace

It would be better to keep Vcc of Driver to about 20 VDC. If the trace length

from output pin of Driver IC to the gate of MOSFET cannot keep at minimum

distance, the width of the tract should be increased to minimize the loop inductance.

For every tiny increase in conductor length between output pin of Driver IC to the

Gate lead of MOSFET, there will be a major increase in rise time. Besides, it will

cause transmission line effect and resultant RFI/EMI [19]. This inductance could also

resonate with parasitic capacitances of MOSFET, making it hard to acquire clean

current waveforms at rise and fall. Every MOSFET also has some inductance and the

lower this value; the better is the switching performance.

17

2.7 MOSFET

MOSFET had been used in most applications of Modern Power Electronics.

In future, we will see more and more applications making use of MOSFET.

MOSFETs can be switched at much higher frequencies with the absence of minority

carrier transport. The limit is imposed by two factors: transit time of electrons across

the drift region and the time required to charge and discharge the input Gate and

„Miller‟ capacitances [20].

Figure 2.7: Symbol and equivalent circuit of a MOSFET

N-Channel MOSFET symbol and an equivalent circuit of MOSFET model

with three inter-junction parasitic capacitances, namely: CGS, CGD and CDS is shown

in Figure 2.7. CGD is referred to the „Miller‟ capacitance and it contributes most to

the switching speed limitation of the MOSFET [20]. Before drain current ID can

begin to flow, CGS needs to be charged to a critical threshold voltage level VGS(th).

Figure 2.8 shows the curve of ID versus VGS for a power MOSFET. It has a slope

which equal to the transconductance, gm. For power MOSFETs, it is

appropriate to consider the relationship to be linear for values of VGS above VGS(th).

The relationship between VGS and ID is parabolic in nature:

[ ] (2.8)

18

Figure 2.8: Transfer characteristics of a power MOSFET

2.7.1 MOSFET Turn-on Phenomena

A MOSFET being turned on by a driver in a clamped inductive load is shown

in Figure 2.9. To initiate the conduction from Drain to Source, CGS of MOSFET

needs to be charged to a critical voltage level VGS(th). A current source with a diode D

connected antiparallel across the inductor represent the clamped inductive load [20].

RGint is the MOSFET intrinsic internal Gate resistance. The inter-junction parametric

capacitances (CGS, CGD and CDS) are connected at their own way. VDD is the dc

voltage supply connected to the Drain of the MOSFET through the clamped

inductive load. Rdr is the output source impedance of the Driver. Vcc with the value

of Vp will supply the Driver and its ground is connected to the common ground of

VDD and then returned to the Source of the MOSFET. There will be an amplified

pulse with the amplitude of Vp at the output terminal of the Driver when a positive

going pulse entering the input terminal of the Driver. The output from Driver is fed

to the Gate of the MOSFET through RGext. RGext is the resistance in series with the

Gate of a MOSFET to control the switching speed of the MOSFET [20].

19

Figure 2.9: A MOSFET being turned on by a driver in a clamped inductive load

Figure 2.10: A MOSFET being turned off by a driver in a clamped inductive load

20

Figure 2.11: MOSFET turn on sequence

MOSFET turn on sequence with variation of different parameters versus time

is shown in Figure 2.11. From t0 to t1, (CGS+CGDl) is exponentially charged until VGS

reaches VGS(th) with a time constant T1. In this time period, Drain voltage remains at

VDD and Drain current, ID has not commenced yet. Between 0 to t1, as VGS rises, IGS

falls exponentially, because it is an RC Circuit. After time t1, as the VGS rises above

VGS(th), MOSFET enters its linear region. At time t1, ID initiates, but the VDS is still

maintain at VDD. However, after t1, ID builds up rapidly. Between t1 to t2, the ID

increases linearly with respect to VGS. At time t2, VGS enters the Miller Plateau level

and VD begins to fall quickly [20].

21

From t2 to t4, VGS and IGS remain at the same value. This is called the Miller

Plateau Region. In this time period, most of the drive current from the driver is used

to discharge the CGD and enhance rapid fall of VDS. The ID will only be restricted by

the external impedance in series with VDD. After t4, VGS begins to exponentially rise

again with a time constant T2. During this time interval the MOSFET gets fully

enhanced and the final value of the VGS will determine the effective RDS(on). When

VGS reaches its final value, VDS attains its lowest value, determined by VDS= IDS x

RDS(on) [20].

2.7.2 MOSFET Turn-off Phenomena

Figure 2.10 shows a MOSFET being turned off by a driver in a clamped

inductive load. The turn-off phenomenon is shown in Figure 2.12. There are two

different decay rates when the output from the Driver drops to zero for turning off

MOSFET. From time 0 to t1, VGS initially decays exponentially at the rate of time

constant T2 but it decays exponentially at the rate of time constant T1 when beyond

T4. The first delay here is required to discharge the CISS capacitance from its initial

value to the Miller Plateau level [20]. From t0 to t1, the gate current is flowing

through CGS and CGD capacitances of MOSFET. During this time interval, ID remains

constant but VDS begins to rise. From t1 to t2, VDS rises from VDS(on) towards its final

off state value of VDS(off) .

From t3 to t4, the VGS begins to fall further below VGS(th). CGS will be

discharge through any external impedance between Gate and Source terminals. The

MOSFET is in its linear region and ID drops rapidly towards zero value. At the

beginning of this interval, the VDS was at its off state value VDS(off) and the MOSFET

will completely turned off at t4 [20].

22

Figure 2.12: MOSFET turn off sequence

2.8 Magnetic Effects on Living Organism

It is proven that magnetic fields do affect the human body in different ways,

and these may be best to be used in therapy [14]. When injury occur, that particular

area is magnetically positive (south pole) and it will become magnetically negative

after few hours (healing occur). North pole magnets have the ability to end the

23

enlargement of growth and contamination while south pole magnets able to arouse

growth of tissue and living systems (bacteria). Hundreds of experiments show that

north pole, the magnetic negative energy has decelerated, controlled and stopped

further development of active cancer site [15]. Due to its characteristic, north pole is

preferred as it has healing effects, relieves pains, reduce inflammations as well as the

infections.

2.9 Applications of Plasma Needle

2.9.1 Dental Applications

Major problem in dentistry are dental cavities. At the moment, laser

technique or traditional method, mechanical drilling can be used to clean the cavities

in teeth. The cleaning process is purely depends on the skill of the dentist. During the

cleaning process, patient might be suffering as both methods involve heating or

vibrations. Heating and vibrations can aggravate the nerve and it can be very painful

to the patient.

Plasma can prevent patient from suffering as it can sterilize the dental cavities

without going through drilling and heating process. This is because the active plasma

species it produces can easily access small irregular cavity and gap. The excess

active species will recombine among themselves or reacting with ambient air and

water molecules once the treatment is completed. Besides, plasma does not cause

major heating to dental pulp yet able to kill the bacteria [10]. Furthermore, cost of

plasma is relatively inexpensive.

24

2.9.2 Plasma Treatment of Mammalian Vascular Cells

Plasma treatment also offers possibility in aiding wound healing. Experiment

between plasma needle and cultured cells were carried out to test for the effects of

electrical properties on cell detachment and necrosis. Mammalian cells that placed

under plasma have the 10s of treatment time for detachment. For short exposure

under plasma yet has the good effect, the layer needed to be thin which is less than

1mm [11]. This has opened the space for cold plasma to be used in healing the

wound.

2.9.3 Cancer Treatment

Nowadays, the only treatment for cancer is either chemotherapy or surgery.

Chemotherapy works against cancer by damaging the nucleus of the cells in the body

that are undergoing the process of division. The cancer-fighting drugs are injected

directly into the body and it will travel around the body to damage and kill the cancer

cells where it has spread. However, it has the side effect of stopping the hair follicles

and skin from dividing.

Plasma treatment can prevent patients from suffering as it does not have the

side effects. Plasma treatment has high-precision on the cells as there will be sharp

boundary line between non-plasma treated region and plasma-treated regions [12].

The high-precision of plasma treatment allows it to be applied directly to the cancer

cells without damaging the surrounding normal cells especially in biomedical

application.

25

CHAPTER 3

RESEARCH METHODOLOGY

3.1 Introduction

This chapter will discuss the methodology used to complete the design of

plasma needle and high frequency power supply for bio-medical applications. Sub

title included in this chapter is methodology procedure, related guidelines and

datasheet, and software used for modelling. Methodology procedure will list out all

necessary steps to finish the design in a simple flow chart. Guidelines and datasheets

that related to the design will be study as well. Software used in this study are

SolidWorks 2011 and Multisim 10.0.

In this study, a novel design of plasma needle will be designed by using

SolidWorks 2011. A ring magnet will be added to the head of the plasma needle and

it is expected to increase the plasma uniformity as well as the healing effect on the

treated area. Multisim 10.0 will be used for power supply circuit simulation. The

circuit will be simulated in Multisim 10.0 before proceed to hardware development

in breadboard and followed by printed circuit board.

26

3.2 Methodology Procedure

The methodology of the design of plasma needle and high frequency power

supply is as summarized in the flow chart below:

Literature Review

Designing and simulation of

Class E amplifier

Designing a plasma needle with

a ring magnet

Hardware development of

plasma needle

Campare the hardware result with the

simulation result

Draw conclusion and

report preparation

27

3.3 Related Guidelines and Datasheets

In order to design the plasma needle and high frequency power supply, the

related journal papers were read and used as references and guidelines. The

specifications of designed plasma needle will be based on the current plasma needle

designed by Eva Stoffels and her team. Same thing goes to the high frequency power

supply as well. The power supply circuit will be modified base on the current circuit.

Besides, related datasheets will be read so that the best components will be chosen in

the circuit. It is very important to choose the correct values and ratings for the

components used in the designed circuit.

3.4 Software Used for Modelling

In this study, SolidWorks 2011 will be used to design the plasma needle

while Multisim 10.0 will be used for class E power amplifier circuit simulation.

3.4.1 SolidWorks 2011

SolidWorks 2011 was chosen as the software to design the plasma needle. It

is a 3D software tools which is more user friendly and easier to learn. It has variety

of user interface tools and capabilities to help designers in creating and editing

models well. Those tools are windows functions, SolidWorks document windows

and function selection and feedback.

28

SolidWorks patent-pending rapid dimensions show the new dimension

placement options and neatly re-arrange current dimensions for selection. The

configuration publisher lets users easily publish a model configurator interface to a

web-based marketplace for 3D parts, assemblies, and other content for selection of

model options. Besides, SolidWorks 2011 also offers improved reference plane

creation methods, sheet metal functionality, weldment performance, component

imaging capabilities, and direct editing tools.

SolidWorks Simulation Premium arms the users with tools to easily validate

design decisions, uncover hidden problems before they affect production and hence

save a lot of money in prototyping. New version also has the event-based motion

simulation, proximity sensors and automatic edge-weld sizing. The overhauled

simulation advisor will guide learners through their first few successful simulations,

shortening the learning curve and adding a layer of protection against errors.

SolidWorks Sustainability software also makes the sustainable design reliable,

reachable and simple. It helps users to govern the carbon footprint, energy

consumption, and air/water impacts in a product design‟s raw material sourcing,

manufacture, use, and disposal. The users also can compare multiple designs at the

same time due to the configuration support in this software. The assembly

visualization tool color-codes parts are based on their total environmental impact.

SolidWorks 2011 was designed to be quicker and more proficiently,

optimized the support for manufacturing and upgraded the collaboration and

visualization. It allows users to stay longer in this software without reboot the

application. On the other hand, sustainability data also can be added to the assembly

visualization tool. In short, SolidWorks 2011 provides more than 200 enhancements

to be more user-friendly so that they will have more expressive experience and

results throughout the entire process.

29

3.4.2 Multisim 10.0

Multisim is an SPICE simulation program used in industry and classroom

teaching. It is the basis of the NI circuits to train a professional circuit designer

through practical application in designing, modelling, and testing of electrical

circuits. Besides, it is also designed for schematic entry, simulation, and finally leads

to implementation and production of PCB layout.

Multisim provides the reliable circuit design for expertise. It keeps improving

to ensure the circuit designers and researchers can move faster to the stage of PCB

production. One of the advantages of circuit design by using this software is the

designers will have the accurate part selection. Multisim has the database of more

than 22,000 components from top semiconductor manufacturers such as Analog

Devices, National Semiconductor, NXP, ON Semiconductor, and Texas Instruments.

Secondly, it can verify the designs by the simulation. The simulation result

can be used for analysis purpose. Multisim has around 20 industry standard SPICE

analyser and 22 measurement instruments for the designer to validate the

performance of the designed circuit. Thirdly, the design circuit can be translated

faster to PCB prototype since the NI Ultiboard layout environment is wholly

integrated with Multisim. It can save the transfer time and ease the designer work.

Lastly, the design prototype also can be validated by LabVIEW which

integrated together with NI Multisim. The relation between real and simulated results

performance can be verified by integrating the Multisim measurements into NI

LabVIEW. This is important to ensure the physical prototype meets the

specifications. Multisim was chosen as simulation software due to the advantages

stated above and more importantly it is free and easy to learn.

30

CHAPTER 4

PLASMA NEEDLE DESIGN

4.1 Introduction

This section will discuss the design of plasma needle. It will discuss about the

components used, the dimension and the description of the model. The plasma needle

will be designed by using SolidWorks 2011 software.

4.2 Modelling Components

Plasma needle has a very simple design. It comprised of a tungsten needle,

ceramic rod, Pyrex tube, cylinder shape plastic, plastic holder, BNC female, washer

and ring magnet. Figure 4.1 shows the components of the designed plasma needle.

The holder is made of plastic so that it will has a lighter weight and be more user-

friendly.

31

Figure 4.1: Components of Plasma Needle

4.3 Modelling Dimensions

Figure 4.2: Dimensions of Plasma Needle

Cylinder Shape

Pyrex Tube

Washer

Tungsten Needle

Ceramic Rod

BNC Female

Ring Magnet

Plastic Holder

32

The designed plasma needle has the length of 118mm with the width of

18mm. it is kept as short and thin as possible so that it is easier for handling. The

plastic holder in the design has the length of 50mm and width of 20mm. One end of

the plastic holder is connected to the BNC female while another end is connected to

the Perspex tube. The gas inlet in the plastic holder has the slope of 450. It is

designed with the slope of 450

so that the gas inlet pipe does not obstruct the

treatment process when plasma needle is being applied in dentistry or other

applications.

Besides, there is a 2mm thick Perspex between the washer and the ring

magnet. The 2mm Perspex is used to fixed the ring magnet at its position. The ring

magnet has the outer diameter (OD) of 18mm and internal diameter (ID) of 12mm.

This ring magnet has to be custom-made as the size is too small and rarely found in

the magnet manufacturing plant.

4.4 Modelling Descriptions

The designed plasma needle has a very simple design. It comprised of a

tungsten needle, ceramic rod, Pyrex tube, cylinder shape plastic, plastic holder, BNC

female, washer and ring magnet. The needle consisted of a tungsten needle (electrode)

of 0.5mm in diameter which is encapsulated in a ceramic rod. The ceramic rod is

used to provide mechanical support as well as electrical insulation for the electrode.

There is also a cylinder shape plastic with holes in the middle of the plasma needle to

provide extra mechanical support to the tungsten and ceramic. The hole in the middle

of cylinder shape plastic is to put the power electrode while the other holes are for

gas helium to go through.

33

Figure 4.3: Plasma Needle Model

The tip of tungsten needle is uncovered 2mm so that the tungsten can mix

with helium to create the plasma. Both electrode and ceramic tube are embedded in

the Pyrex tube. The Pyrex tube is used to insulate the needle to prevent a discharge

along the pin. The tube is protrudes from the plastic holder. The plastic holder is

designed with a gas inlet hence the gas helium can flow in from the gas inlet.

Another end of the plastic holder is threaded inside so that it can grip the BNC

female tightly. BNC will be connected to the power supply via the coaxial cable.

The novel design is made by adding a 10mm ring magnet at the tip of the

plasma needle. A washer is added so that the ring magnet can be attracted to it. The

ring magnet with different magnetic field strength (Tesla) will be tested to find for

the best magnetic field strength in creating the uniform plasma besides having the

healing effect on the treated area. The washer will only be used in the experiment

stage where it can be removed after the best magnetic field strength was found. The

ring magnet can be glued directly to the Perspex tube after confirm the magnetic

field strength of the ring magnet.

RF Signal Ring Magnet

Gas Inlet

34

Figure 4.4: The Pyrex tube moved 5mm into the plastic holder

The Perspex tube which protrudes from the plastic holder can move 5mm into

the plastic holder as shown in Figure 4.4. This is specially design for sterilizing

purpose. Even though the plasma kills most of the bacteria, the plasma needle needs

to be be sterilized after the treatment. Figure 4.5 shows the designed plasma needle

after rendered by using SolidWorks 2011.

Figure 4.5: Plasma Needle after Rendered

Ring Magnet

Needle Tip

Gas Inlet

35

CHAPTER 5

HIGH FREQUENCY POWER SUPPLY

5.1 Introduction

This section discusses the design of high frequency power supply for bio-

medical applications. This research aims to design a high frequency power supply

which can generate a voltage up to 200V with the frequency of 13.56MHz. The

design was started with the software simulation and followed by hardware

development. The simulation and hardware development will be discussed in detail.

5.2 Simulation Development

The design of high frequency power supply was started with software

simulation. This is the step to verify whether the design circuit is functioning well

before continue with hardware development. In this research, Multisim 10.0 was

used as the simulation software.

36

5.2.1 Simulation of Modified Class E Amplifier

Figure 5.1: Modified Class E Amplifier Simulation Circuit

Figure 5.1 shows the modified Class E amplifier simulation circuit. This

circuit is very simple yet able to generate high voltage and high frequency in

generating non-thermal plasma. The component L1 is functioning as a choke

inductor for filtering. It allows only DC signal to pass through. The RLC load

consisted of R1, L3 and C2. L2 and C1 were combined to form the resonant circuit

with the frequency of 13.56MHz. Their value can be adjusted based on the input

frequency by using the formula

√ (5.1)

The voltage range from 200Vpp to 600Vpp is needed in generating non-

thermal plasma. The voltage is inversely proportional to the required helium flow

rate. The higher the voltage the lower the gas helium required to generate the plasma.

When the gas helium is lower, the cost of generating the plasma is lowered down as

well.

VCT VCR VO

37

According to datasheet RF MOSFET DE275X2-102N06A, it has the VGS(th)

range from 2.5V to 5.5V. Hence, the 5Vp square wave input signal was used to drive

the MOSFET in the simulation. The 5Vp square wave signal was directly enter into

MOSFET without go through a driver. However, there will be a MOSFET driver,

DEIC515 in the real circuit to drive the MOSFET.

In the simulation circuit, the function generator with 5Vp square wave was

used to replace the MAX038. MAX038 is the signal generator IC which able to

generate the waveform frequency up to 20MHz. The input voltage of 20V was used

in the modified Class E amplifier simulation circuit. The 20V input voltage was

connected to the 33µH choke inductor.

The Simulated Tektronix Oscilloscope was used in the simulation software to

visualize the output from the circuit. Since the Simulated Tektronix Oscilloscope has

four probes, it can have four outputs at the same time. In this simulation, only two

output signals were shown at the same time.

5.3 Hardware Development

Hardware development was conducted after the software simulation and

components selection. It was started with frequency waveform generator, MAX038,

followed by Driver DEIC 515 and MOSFET DE275X2-106N06A.

38

5.3.1 MAX038 Frequency Waveform Generator

Figure 5.2: MAX038 Frequency Waveform Generator Circuit

Figure 5.2 shows the MAX038 waveform generator circuit. MAX038 can

generate square wave, sine wave and triangle wave. The input to A0 ( pin 3 ) and A1

( pin 4 ) will determine type of waveforms being generated. The MOSFET needs

square wave for its input signal. Hence, both inputs to A0 and A1 were set to 0 to

have the square wave output signal. CF was fixed at 33pF and the frequency of the

output waveform can be adjusted by varying the 20kΩ potentiometer, RIN.

39

Figure 5.3: MAX038 Hardware Circuit

The +5V was connected to pin 17 while -5V was connected to pin 20 of

MAX038. The ±5V was supplied from the regulated DC power supply. The 4.7µH

capacitors were used to reduce noise and stabilize the input voltage to MAX038. The

output from MAX038 will be connected to the input of MOSFET Driver, DEIC515.

5.3.2 DEIC515 MOSFET Driver Circuit

Figure 5.4: DEIC515 MOSFET Driver Circuit

RIN, 20kΩ

MAX038

4.7µ

F

12kΩ

1nF Output

-5V

+5V

40

DEIC515 is a 15A low side ultrafast RF MOSFET driver. It can drive the

MOSFET with the frequency up to 45MHz. However, a very large transient will be

created in DEIC515 due to the high currents and high speeds. L1 as shown in Figure

5.4 is the simple tri-filar winding on a small ferrite core. It is the common mode

choke used at the DEIC515 input to avoid false triggering by directing the input

signals to follow the internal die potential changes.

The input and ground of the square wave were connected to the common

mode choke before entering IN and INGND of the DEIC515. The input voltage, VCC

for DEIC515 was 15V. The input square wave signal has the amplitude of 5Vp. The

output from DEIC 515 was connected to MOSFET DE275X2-102N06A input.

Figure 5.5: DEIC515 Hardware Circuit

5.3.3 DEIC515 and DE275X2-102N06A MOSFET Circuit

Figure 5.6 shows the Driver DEIC515 and MOSFET DE275X2-102N06A

circuit on breadboard. The driver DEIC515 was used to drive the MOSFET at

frequency of 5MHz and 8MHz. The output from DEIC515 was connected to the

Gate of the MOSFET. The input voltage, VDD was set at 15V and connected to the

Output 10µF

10µF VCC

DEIC515

INVCC

INGND

IN

Ground

41

Drain of the MOSFET. The output from MOSFET was attached to the Class E

amplifier circuit.

Figure 5.6: DEIC515 and MOSFET Hardware Circuit

DEIC515

DE275X2-102N06A

Output

VDD Ground

42

CHAPTER 6

RESULTS AND DISCUSSIONS

6.1 Introduction

This section discusses the simulation and the experimental results in this

research. It will start with the simulation result of modified Class E amplifier and

follow by the MAX038, DEIC515 and DE275X2-102N06A hardware results. The

output from the MOSFET, DE275X2-102N06A will be combined with the class E

amplifier to get the high voltage and high frequency output. The results obtained

from the experiment will be compared with the simulation results.

6.2 Simulation Results For Modified Class E Amplifier

Figure 6.1 and 6.2 show the simulated waveforms of VCT, VCR, Vo and Vin.

The simulated results as shown in Figure 6.1 and 6.2 prove that the modified Class E

amplifier able to generate an output voltage of 505Vpp at the frequency of

13.56MHz with an input voltage of 20V. The output voltage is high enough to

43

generate non-thermal plasma after mix with the helium gas. VCT is the simulated

waveform that obtained from the Drain of the MOSFET. It was amplified from 15V

to 99.6Vpp.

Figure 6.1: Simulated waveform of VCT and VCR

Figure 6.2: Simulated waveform of Vin and Vo

All the components are assumed ideal in the simulation. In reality, many

factors will affect the output of the circuit. The output voltage will be lower than the

simulation result in real experiment.

VCT

VCR

Vo

Vin

44

6.3 Hardware Results

The results of MAX038 frequency waveform generator, DEIC515 MOSFET

Driver as well as DE275X2-102N06A MOSFET will be shown in this section.

6.3.1 MAX038 Frequency Waveform Generator

Figure 6.3(a), (b) and (c) show the output signals from MAX038 at frequency

of 5MHz, 10MHz and 13.75MHz. The frequency was varied by adjusting the 20kΩ

potentiometer. The results show that MAX038 can generate square wave at high

frequency. It can be used as the signal generator to replace the function generator in

generating input signal for MOSFET. This will definitely reduce the cost of the power

supply since MAX038 is cheaper than a function generator.

(a)

45

(b)

(c)

Figure 6.3: Output of MAX038 at (a) 5MHz, (b) 10MHz and (c) 13.75MHz

6.3.2 DEIC515 MOSFET Driver

Figure 6.4(a), (b) and (c) show the DEIC515 MOSFET Driver output signal at

frequency of 1MHz, 5MHz and 10MHz. It can be observed that the output from the

Driver has the amplitude of 14Vpp at the frequency of 1MHz. At frequency of 5MHz,

the amplitude was reduced to 7Vpp. The amplitude was decreased to 3.5Vpp at

frequency of 10MHz. However, the square shape still can be observed although the

46

frequency increased to 10MHz. This results show that Driver DEIC515 has the ability

to drive the signal at high frequency.

(a)

(b)

47

(c)

Figure 6.4: Output of DEIC515 at (a) 1MHz, (b) 5MHz and (c) 10MHz

The problem faced was the signal output voltages were decreasing when the

frequency increased. The voltage at the voltage generator also decreased when the

frequency of the input signal increased. There were voltage drop along the circuit

when high frequency being applied to the Driver. The circuit path was inductive and

the reactance, XL= jwL. When f=13.56MHz, the reactance was very large. At this

time, it is believed that most of the supply voltage did not pass through the Driver but

go through the capacitor that connected between the VCC and the ground.

6.3.3 DEIC515 and DE275X2-102N06A MOSFET

Figure 6.5 shows the output signals of MOSFET DE275X2-102N06A being

drove by DEIC515. The output signal from MOSFET has the output voltage of

2.53Vpp at frequency of 5MHz. At frequency of 8MHz, the signal has output voltage

of 1.25Vpp.

48

(a)

(b)

Figure 6.5: Output of DE275X2-102N06A at (a) 5MHz and (b) 8MHz

From the results, it can be observed that the voltage decrease when the

frequency was being increased. It is believed that the MOSFET is not ON during the

experiment since no amplification occurred. The voltage drop along the circuit paths

need to be tackled in order to solve this problem.

49

CHAPTER 7

CONCLUSION & RECOMMENDATION

7.1 Introduction

In this chapter, the researcher will restate the objectives and make the

conclusion based on the results. Besides, recommendation for future research also

will be indicated.

7.2 Conclusion

This research had presented the design of plasma needle and high frequency

power supply for bio-medical applications. First objective of the research to design a

plasma needle was achieved successfully. SolidWorks 2011 software was used to

design the plasma needle. The modelling components, dimensions and descriptions

were discussed in chapter 4 previously. The newly design plasma needle is expected

to improve the plasma uniformity as well as the healing effects on the treated area. It

will be lighter and easier for handling with diverse design and material used.

50

The second objective of the research to design the high frequency power

supply was partially achieved. The simulation of the power supply circuit was

simulated successfully where it managed to produce the high voltage at high

frequency. When it comes to the hardware development, the output at both Driver

and MOSFET were not similar to the simulation result. The DEIC515 and

DE275X2-102N06A were able to drive the input signal at high frequency. However,

the voltage dropping at high frequency needs to be settled before proceed to the

amplifier circuit stage.

Current method of high frequency power supply is very expensive since it

consisted of commercialize function generator and research amplifier. These

equipments are specially used in laboratory for researching and studying purpose.

The research and improvement on the high frequency power supply must be carried

on in order to develop a cheaper yet simpler power supply for future bio-medical

applications.

7.3 Recommendation

Based on the research of this study, here are several recommendations for

future work to improve the high frequency power supply as well as the plasma

needle in bio-medical applications:

1. MAX038 is able to generate the output waveform at high frequency. It can be

used to replace the function generator in the power supply circuit. However,

MAX038 is no longer manufactured by MAXIM and it is hard to buy a large

amount of MAX038 in the market. Due to this reason, an alternative for

MAX038 should be found.

51

2. For RF design, the high speed switching and losses always become the problem

in circuit design. Using DEIC515 and MOSFET DE275X2-102N06A in the

designed circuit was the right one. It has an excellent thermal transfer and able to

optimize the switching speed of the output waveform. Hence, the RF Driver and

RF MOSFET should be used in the future research.

3. During the hardware development, the RF components should be soldered

directly on the PCB instead of testing on the breadboard. The circuit paths also

must be kept as short as possible. These ways are taken to reduce the noises and

power losses during the high speed switching.

4. The introduction of ring magnet is expected to improve the plasma uniformity as

well as the healing effect at the treated area. Therefore, the designed plasma

needle should be fabricated in order to investigate its effects. The most suitable

magnetic field strength can be determined after the experiment.

52

REFERENCES

1. M K Boudam, M. M., B Saoudi, C Popovici, N Gherardi and F Massines

(2006). Bacterial Spore Inactivation by Atmospheric-Pressure Plasma In The

Presence or Absence of UV Photons as Obtained with the Same Gas Mixture.

2. Kieft, I. E., M. Kurdi, et al. (2006). Reattachment and Apoptosis After

Plasma-Needle Treatment of Cultured Cells. Plasma Science, IEEE

Transactions on 34(4): 1331-1336.

3. Roxana Silvia Tipa, G. M. W. K. (2011). Plasma-Stimulated Wound Healing.

IEEE Transactions On Plasma Science 39(11): 2978-2979.

4. Laroussi, M. (2009). Low-Temperature Plasmas for Medicine? Plasma

Science, IEEE Transactions on 37(6): 714-725.

5. E Stoffels, A. J. F., W W Stoffels and G M W Kroesen (2002). Plasma needle:

a non-destructive atmospheric plasma source for fine surface treatment of

(bio)materials. Plasma Sources Science And Technology 11: 383-388.

6. H Conrads, M. S. (2000) Plasma generation and plasma sources. 9, 441-454

7. I.E.Kieft (2005). Plasma Needle: exploring biomedical applications of non-

thermal plasmas, Printservice Technische Universiteit Eindhoven: 153.

8. Moisan, M., J. Barbeau, et al. (2001). Low-temperature sterilization using gas

plasmas: a review of the experiments and an analysis of the inactivation

mechanisms. International Journal of Pharmaceutics 226(1–2): 1-21.

53

9. M K Boudam, M. M., B Saoudi, C Popovici, N Gherardi and F Massines

(2006). Bacterial spore inactivation by atmospheric-pressure plasma in the

presence or absence of UV photons as obtained with the same gas mixture.

10. Sladek, R. E. J., E. Stoffels, et al. (2004). Plasma Treatment of Dental

Cavities: A Feasibility Study. Plasma Science, IEEE Transactions on 32(4):

1540-1543.

11. Ingrid E. Kieft, D. D., Anton J.M. Roks and Eva Stoffels (2005). Plasma

Treatment of Mammalian Vascular Cells: A Quantitative Description. IEEE

Transactions on Plasma Science 33(2): 771-775.

12. D. Kim, B. G., D.B. Kim, W. Choe and J.H. Shin (2009). A Feasibility Study

for the Cancer Therapy Using Cold Plasma. ICBME: 355-357.

13. Sokal, N. O. (Jan/Feb 2001) Class-E RF Power Amplifiers. QEX: 9-20.

14. Rosendo Peña-Eguiluz, M., IEEE, José Arturo Pérez-Martínez, Régulo

López-Callejas, and J. S.-P. Antonio Mercado-Cabrera, Blanca Aguilar-

Uscanga, Arturo E. Muñoz-Castro, Raúl Valencia-Alvarado, Samuel R.

Barocio-Delgado, Benjamín G. Rodríguez-Méndez, and Aníbal de la Piedad-

Beneitez (2010). Analysis and Application of a Parallel E-Class Amplifier as

RF Plasma Source. IEEE Transactions on Plasma Science 38(10).

15. Jose A. Perez-Martinex, R. P.-E., Regulo Lopez-Callejas, Antonio Mercado-

Cabrera, Raul Valencia Alvarado, Samuel R. Barocio, Anibal de la Piedad-

Beneitez (2008). Power Supply for Plasma Torches Based on a Class-E

Amplifier Configuration.

16. Sadafi, H. A. (1998). The Therapeutic Applications of Pulsed and Static

Magnetic Fields. 2nd International Conference on Bioelectromagnetism.

Melbourne, Australia.

54

17. Albert Roy Davis and Walter C. Rawls, J. (1988). The Magnetic Effect and

Magnetism and Its Effects on the Living System, Exposition Press.

18. Ven, G. v. d. (2006). BEP: Design of a guiding mechanism for the plasma

needle”, Technische Universiteit Eindhoven.

19. Abhijit D. Pathak, S. O. (2003). Unique MOSFET/IGBT Drivers and Their

Applications in Future Power Electronics Systems. Power Electronics and

Drive Systems, PEDS 2003. 1: 85-88.

20. Abhijit D. Pathak (2001). MOSFET/IGBT Drivers, Theory and Applications,

IXYS Corporation.

21. Zirnheld, J. L., S. N. Zucker, et al. (2010). Nonthermal Plasma Needle:

Development and Targeting of Melanoma Cells. Plasma Science, IEEE

Transactions on 38(4): 948-952.

22. Lo Keat How (2011). Modeling And Design of Plasma Needle Supply. IVAT.

Johor, Universiti Teknologi Malaysia. Bachelor of Engineering (Electrical).

55

APPENDIX A

Apparatus for Hardware Experiment

Regulated DC Power Supply, PSM 2/5A

Function Generator, Tektronix AFG 3021B

56

Oscilloscope, LeCray LT344L

Fluke Multimeter

57

APPENDIX B

Datasheet

________________General DescriptionThe MAX038 is a high-frequency, precision functiongenerator producing accurate, high-frequency triangle, sawtooth, sine, square, and pulse waveforms with aminimum of external components. The output frequencycan be controlled over a frequency range of 0.1Hz to20MHz by an internal 2.5V bandgap voltage reference and an external resistor and capacitor. Theduty cycle can be varied over a wide range by applyinga ±2.3V control signal, facilitating pulse-width modula-tion and the generation of sawtooth waveforms.Frequency modulation and frequency sweeping areachieved in the same way. The duty cycle and frequency controls are independent.

Sine, square, or triangle waveforms can be selected atthe output by setting the appropriate code at two TTL-compatible select pins. The output signal for allwaveforms is a 2VP-P signal that is symmetrical aroundground. The low-impedance output can drive up to ±20mA.

The TTL-compatible SYNC output from the internaloscillator maintains a 50% duty cycle—regardless ofthe duty cycle of the other waveforms—to synchronizeother devices in the system. The internal oscillator canbe synchronized to an external TTL clock connected to PDI.

________________________ApplicationsPrecision Function Generators

Voltage-Controlled Oscillators

Frequency Modulators

Pulse-Width Modulators

Phase-Locked Loops

Frequency Synthesizer

FSK Generator—Sine and Square Waves

____________________________Features♦ 0.1Hz to 20MHz Operating Frequency Range

♦ Triangle, Sawtooth, Sine, Square, and PulseWaveforms

♦ Independent Frequency and Duty-CycleAdjustments

♦ 350 to 1 Frequency Sweep Range

♦ 15% to 85% Variable Duty Cycle

♦ Low-Impedance Output Buffer: 0.1Ω

♦ Low 200ppm/°C Temperature Drift

_______________Ordering Information

MA

X0

38

High-Frequency Waveform Generator

________________________________________________________________ Maxim Integrated Products 1

20

19

18

17

16

15

14

13

12

11

1

2

3

4

5

6

7

8

9

10

V-

OUT

GND

V+A1

A0

GND

REF

TOP VIEW

MAX038

DV+

DGND

SYNC

PDIFADJ

DADJ

GND

COSC

PDO

GNDIIN

GND

DIP/SO

___________________ Pin Configuration

19-0266; Rev 5; 2/04

EVALUATION KIT

AVAILABLE

For pricing, delivery, and ordering information, please contact Maxim/Dallas Direct! at 1-888-629-4642, or visit Maxim’s website at www.maxim-ic.com.

PART TEMP RANGE PIN-PACKAGE

MAX038CPP 0°C to +70°C 20 Plastic DIP

MAX038CWP 0°C to +70°C 20 SO

MAX038C/D 0°C to +70°C DiceMAX038EPP* -40°C to +85°C 20 Plastic DIPMAX038EWP* -40°C to +85°C 20 SO

*Contact factory prior to design.

MA

X0

38

High-Frequency Waveform Generator

2 _______________________________________________________________________________________

ABSOLUTE MAXIMUM RATINGS

ELECTRICAL CHARACTERISTICS(Circuit of Figure 1, GND = DGND = 0V, V+ = DV+ = 5V, V- = -5V, VDADJ = VFADJ = VPDI = VPDO = 0V, CF = 100pF,RIN = 25kΩ, RL = 1kΩ, CL = 20pF, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25°C.)

Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent damage to the device. These are stress ratings only, and functionaloperation of the device at these or any other conditions beyond those indicated in the operational sections of the specifications is not implied. Exposure toabsolute maximum rating conditions for extended periods may affect device reliability.

PARAMETER SYMBOL MIN TYP MAX UNITS

Frequency TemperatureCoefficient

∆Fo/°C

200ppm/°C

600

IIN Offset Voltage VIN ±1.0 ±2.0 mV

Frequency ProgrammingCurrent

IIN1.25 375

µA

(∆Fo/Fo)∆V+

±0.4 ±2.00Frequency Power-SupplyRejection (∆Fo/Fo)

∆V-±0.2 ±1.00

%/V

Output Peak-to-Peak Symmetry VOUT ±4 mV

Maximum Operating Frequency Fo 20.0 40.0 MHz

2.50 750

Output Resistance ROUT 0.1 0.2 ΩOutput Short-Circuit Current IOUT 40 mA

Amplitude VOUT 1.9 2.0 2.1 VP-P

Rise Time tR 12 ns

Fall Time tF 12 ns

Duty Cycle dc 47 50 53 %

Amplitude VOUT 1.9 2.0 2.1 VP-P

Nonlinearity 0.5 %

Duty Cycle dc 47 50 53 %

CONDITIONS

VFADJ = -3V

VFADJ = 0V

VFADJ = -3V

V- = -5V, V+ = 4.75V to 5.25V

V+ = 5V, V- = -4.75V to -5.25V

Short circuit to GND

10% to 90%

90% to 10%

VDADJ = 0V, dc = tON/t x 100%

CF ≤ 15pF, IIN = 500µA

VFADJ = 0V

FO = 100kHz, 5% to 95%

VDADJ = 0V (Note 1)

V+ to GND................................................................-0.3V to +6VDV+ to DGND...........................................................-0.3V to +6VV- to GND .................................................................+0.3V to -6VPin Voltages

IIN, FADJ, DADJ, PDO .....................(V- - 0.3V) to (V+ + 0.3V)COSC .....................................................................+0.3V to V-A0, A1, PDI, SYNC, REF.........................................-0.3V to V+GND to DGND ................................................................±0.3V

Maximum Current into Any Pin .........................................±50mAOUT, REF Short-Circuit Duration to GND, V+, V- ...................30s

Continuous Power Dissipation (TA = +70°C)Plastic DIP (derate 11.11mW/°C above +70°C) ..........889mWSO (derate 10.00mW/°C above +70°C).......................800mWCERDIP (derate 11.11mW/°C above +70°C)...............889mW

Operating Temperature RangesMAX038C_ _ .......................................................0°C to +70°CMAX038E_ _ ....................................................-40°C to +85°C

Maximum Junction Temperature .....................................+150°CStorage Temperature Range .............................-65°C to +150°CLead Temperature (soldering, 10s) .................................+300°C

Amplitude VOUT 1.9 2.0 2.1 VP-P

CF = 1000pF, FO = 100kHzTHD %

Fo/°C

FREQUENCY CHARACTERISTICS

OUTPUT AMPLIFIER (applies to all waveforms)

SQUARE-WAVE OUTPUT (RL = 100Ω)

TRIANGLE-WAVE OUTPUT (RL = 100Ω)

SINE-WAVE OUTPUT (RL = 100Ω)

Total Harmonic Distortion 2.0

MA

X0

38

High-Frequency Waveform Generator

_______________________________________________________________________________________ 3

ELECTRICAL CHARACTERISTICS (continued)(Circuit of Figure 1, GND = DGND = 0V, V+ = DV+ = 5V, V- = -5V, VDADJ = VFADJ = VPDI = VPDO = 0V, CF = 100pF,RIN = 25kΩ, RL = 1kΩ, CL = 20pF, TA = TMIN to TMAX, unless otherwise noted. Typical values are at TA = +25°C.)

Note 1: Guaranteed by duty-cycle test on square wave.Note 2: VREF is independent of V-.

PARAMETER

DADJ Nonlinearity

SYMBOL MIN TYP MAX

dc/VFADJ 2 4

UNITS

%

Duty Cycle dcSYNC 50 %Fall Time tF 10 nsRise Time tR 10 ns

Change in Output Frequencywith DADJ

DADJ Input Current IDADJ 190 250 320 µADADJ Voltage Range VDADJ ±2.3 V

Fo/VDADJ ±2.5 ±8 %

Duty-Cycle Adjustment Range dc 15 85 %

Maximum DADJ ModulatingFrequency

FDC 2 MHz

Output Low Voltage

FADJ Input Current IFADJ 190 250 320 µAFADJ Voltage Range VFADJ ±2.4 VFrequency Sweep Range

VOL 0.3 0.4 V

Fo ±70 %FM Nonlinearity with FADJ

Output High Voltage

Fo/VFADJ ±0.2 %

VOH 2.8 3.5 V

Change in Duty Cycle with FADJ dc/VFADJ ±2 %

Output Voltage VREF 2.48 2.50 2.52 V

CONDITIONS

-2V ≤ VDADJ ≤ +2V

90% to 10%, RL = 3kΩ, CL = 15pF10% to 90%, RL = 3kΩ, CL = 15pF

-2V ≤ VDADJ ≤ +2V

-2.3V ≤ VDADJ ≤ +2.3V

ISINK = 3.2mA

-2.4V ≤ VFADJ ≤ +2.4V-2V ≤ VFADJ ≤ +2V

ISOURCE = 400µA

-2V ≤ VFADJ ≤ +2V

IREF = 0

Temperature Coefficient VREF/°C 20 ppm/°C0mA ≤ IREF ≤ 4mA (source) 1 2

Load Regulation VREF/IREF -100µA ≤ IREF ≤ 0µA (sink) 1 4mV/mA

Line Regulation VREF/V+ 4.75V ≤ V+ ≤ 5.25V (Note 2) 1 2 mV/V

Input Low Voltage VIL 0.8 V

Input High Voltage VIH 2.4 V

Input Current (A0, A1) IIL, IIH VA0, VA1 = VIL, VIH ±5 µA

Input Current (PDI) IIL, IIH VPDI = VIL, VIH ±25 µA

Positive Supply Voltage V+ 4.75 5.25 V

SYNC Supply Voltage DV+ 4.75 5.25 V

Negative Supply Voltage V- -4.75 -5.25 V

Positive Supply Current I+ 35 45 mA

SYNC Supply Current IDV+ 1 2 mA

Negative Supply Current I- 45 55 mA

Maximum FADJ ModulatingFrequency FF 2 MHz

SYNC OUTPUT

DUTY-CYCLE ADJUSTMENT (DADJ)

FREQUENCY ADJUSTMENT (FADJ)

VOLTAGE REFERENCE

LOGIC INPUTS (A0, A1, PDI)

POWER SUPPLY

MA

X0

38

High-Frequency Waveform Generator

4 _______________________________________________________________________________________

__________________________________________Typical Operating Characteristics(Circuit of Figure 1, V+ = DV+ = 5V, V- = -5V, VDADJ = VFADJ = VPDI = VPDO = 0V, RL = 1kΩ, CL = 20pF, TA = +25°C, unless otherwise noted.)

0.11 100 1000

OUTPUT FREQUENCYvs. IIN CURRENT

10

100

MAX

038-

08

IIN CURRENT (µA)

OUTP

UT F

REQU

ENCY

(Hz)

10

1

1k

10k

100k

1M

10M

100M

100µF47µF

10µF

3.3µF

1µF

100nF

33nF

3.3nF

330pF

100pF

33pF

1.0

0-3 2

NORMALIZED OUTPUT FREQUENCYvs. FADJ VOLTAGE

0.2

0.8

MAX

038-

09

VFADJ (V)

F OUT

NOR

MAL

IZED

0

0.4

-2 -1 1

0.6

3

1.2

1.4

1.6

1.8

2.0

IIN = 100µA, COSC = 1000pF

0.85

NORMALIZED OUTPUT FREQUENCYvs. DADJ VOLTAGE

0.90

1.10

MAX

038-

17

DADJ (V)

NORM

ALIZ

ED O

UTPU

T FR

EQUE

NCY

1.00

0.95

1.05

IIN = 10µA

IIN = 25µA

IIN = 50µA

IIN = 100µA

IIN = 250µA

IIN = 500µA

2.0

-2.5-2.0 -1.0 1.0 2.5

DUTY-CYCLE LINEARITYvs. DADJ VOLTAGE

-2.0

1.0

MAX

038-

18

DADJ (V)

DUTY

-CYC

LE L

INEA

RITY

ERR

OR (%

)

0 1.5

0

-1.0

-1.5

-0.5

0.5

1.5

IIN = 10µAIIN = 25µA

IIN = 50µA

IIN = 100µA

IIN = 250µA

IIN = 500µA

60

0-3 2

DUTY CYCLE vs. DADJ VOLTAGE

10

50

MAX

038-

16B

DADJ (V)

DUTY

CYC

LE (%

)

0

30

20

-2 -1 1

40

70

80

90

100

3

IIN = 200µA

MA

X0

38

High-Frequency Waveform Generator

_______________________________________________________________________________________ 5

SINE-WAVE OUTPUT (50Hz)

TOP: OUTPUT 50Hz = FoBOTTOM: SYNCIIN = 50µACF = 1µF

TRIANGLE-WAVE OUTPUT (50Hz)

TOP: OUTPUT 50Hz = FoBOTTOM: SYNCIIN = 50µACF = 1µF

SQUARE-WAVE OUTPUT (50Hz)

TOP: OUTPUT 50Hz = FoBOTTOM: SYNCIIN = 50µACF = 1µF

SINE-WAVE OUTPUT (20MHz)

IIN = 400µACF = 20pF

_____________________________Typical Operating Characteristics (continued)(Circuit of Figure 1, V+ = DV+ = 5V, V- = -5V, VDADJ = VFADJ = VPDI = VPDO = 0V, RL = 1kΩ, CL = 20pF, TA = +25°C, unless otherwise noted.)

TRIANGLE-WAVE OUTPUT (20MHz)

IIN = 400µACF = 20pF

FREQUENCY (Hz)

1M100k10k1k100 10M

FREQUENCY (Hz)

1M100k10k1k100 10M

SINE WAVE THD vs. FREQUENCY

MAX

038

toc0

1

FREQUENCY (Hz)

THD

(%)

1M100k10k1k

1

2

3

4

5

6

7

0100 10M

MA

X0

38

High-Frequency Waveform Generator

6 _______________________________________________________________________________________

_____________________________Typical Operating Characteristics (continued)(Circuit of Figure 1, V+ = DV+ = 5V, V- = -5V, VDADJ = VFADJ = VPDI = VPDO = 0V, RL = 1kΩ, CL = 20pF, TA = +25°C, unless otherwise noted.)

FREQUENCY MODULATION USING FADJ

TOP: OUTPUTBOTTOM: FADJ

0.5V

0V

-0.5V

FREQUENCY MODULATION USING IIN

TOP: OUTPUTBOTTOM: IIN

FREQUENCY MODULATION USING IIN

TOP: OUTPUTBOTTOM: IIN

PULSE-WIDTH MODULATION USING DADJ

TOP: SQUARE-WAVE OUT, 2VP-PBOTTOM: VDADJ, -2V to +2.3V

+1V

0V

-1V

+2V

0V

-2V

SQUARE-WAVE OUTPUT (20MHz)

IIN = 400µACF = 20pF

MA

X0

38

High-Frequency Waveform Generator

_______________________________________________________________________________________ 7

______________________________________________________________Pin Description

*The five GND pins are not internally connected. Connect all five GND pins to a quiet ground close to the device. A ground plane isrecommended (see Layout Considerations).

0

-1000 20 60 100

OUTPUT SPECTRUM, SINE WAVE(Fo = 11.5MHz)

-80

-20

MAX

038-

12A

FREQUENCY (MHz)

ATTE

NUAT

ION

(dB)

40 80

-40

-60

-10

-30

-50

-70

-90

10 30 50 70 90

RIN = 15kΩ (VIN = 2.5V), CF = 20pF, VDADJ = 40mV, VFADJ = -3V

0

-1000 10 30 50

OUTPUT SPECTRUM, SINE WAVE(Fo = 5.9kHz)

-80

-20

MAX

038

12B

FREQUENCY (kHz)AT

TENU

ATIO

N (d

B)20 40

-40

-60

-10

-30

-50

-70

-90

5 15 25 35 45

RIN = 51kΩ (VIN = 2.5V), CF = 0.01µF, VDADJ = 50mV, VFADJ = 0V

_____________________________Typical Operating Characteristics (continued)(Circuit of Figure 1, V+ = DV+ = 5V, V- = -5V, VDADJ = VFADJ = VPDI = VPDO = 0V, RL = 1kΩ, CL = 20pF, TA = +25°C, unless otherwise noted.)

-5V supply inputV-20

Sine, square, or triangle outputOUT19

+5V supply inputV+17

Digital +5V supply input. Can be left open if SYNC is not used.DV+16

Digital groundDGND15

TTL/CMOS-compatible output, referenced between DGND and DV+. Permits the internal oscillator to besynchronized with an external signal. Leave open if unused.

SYNC14

Current input for frequency controlIIN10

Phase detector output. Connect to GND if phase detector is not used.PDO12

Phase detector reference clock input. Connect to GND if phase detector is not used.PDI13

External capacitor connectionCOSC5

Duty-cycle adjust inputDADJ7

Frequency adjust inputFADJ8

Waveform selection input; TTL/CMOS compatibleA14

Waveform selection input; TTL/CMOS compatibleA03

PIN

Ground*GND2, 6, 9,11, 18

2.50V bandgap voltage reference outputREF1

FUNCTIONNAME

MA

X0

38

_______________Detailed DescriptionThe MAX038 is a high-frequency function generatorthat produces low-distortion sine, triangle, sawtooth, orsquare (pulse) waveforms at frequencies from less than1Hz to 20MHz or more, using a minimum of externalcomponents. Frequency and duty cycle can be inde-pendently controlled by programming the current, volt-age, or resistance. The desired output waveform isselected under logic control by setting the appropriatecode at the A0 and A1 inputs. A SYNC output andphase detector are included to simplify designs requir-ing tracking to an external signal source.

The MAX038 operates with ±5V ±5% power supplies.The basic oscillator is a relaxation type that operates byalternately charging and discharging a capacitor, CF,

with constant currents, simultaneously producing a tri-angle wave and a square wave (Figure 1). The charg-ing and discharging currents are controlled by the cur-rent flowing into IIN, and are modulated by the voltagesapplied to FADJ and DADJ. The current into IIN can bevaried from 2µA to 750µA, producing more than twodecades of frequency for any value of CF. Applying±2.4V to FADJ changes the nominal frequency (withVFADJ = 0V) by ±70%; this procedure can be used forfine control.

Duty cycle (the percentage of time that the output wave-form is positive) can be controlled from 10% to 90% byapplying ±2.3V to DADJ. This voltage changes the CFcharging and discharging current ratio while maintainingnearly constant frequency.

High-Frequency Waveform Generator

8 _______________________________________________________________________________________

MAX038

OSCILLATOR

OSCILLATORCURRENT

GENERATOR

2.5VVOLTAGE

REFERENCE

OSC B

OSC ATRIANGLE

SINESHAPER

COMPARATOR

COMPARATOR

PHASEDETECTOR

MUX

COSC

GND

5

6CF

8

7

10

FADJ

DADJ

IIN

REF1

1720

2, 9, 11, 18

V+V-

GND

RF RD RIN

+5V

-5V

-250µA

SINE

TRIANGLE

SQUARE

A0 A1

OUT

SYNC

PDO

PDI

19

14

12

13

RL CL

3 4

DGND DV+15 16

+5V

*

= SIGNAL DIRECTION, NOT POLARITY

= BYPASS CAPACITORS ARE 1µF CERAMIC OR 1µF ELECTROLYTIC IN PARALLEL WITH 1nF CERAMIC.

*

*Figure 1. Block Diagram and Basic Operating Circuit

A stable 2.5V reference voltage, REF, allows simpledetermination of IIN, FADJ, or DADJ with fixed resistors,and permits adjustable operation when potentiometersare connected from each of these inputs to REF. FADJand/or DADJ can be grounded, producing the nominalfrequency with a 50% duty cycle.

The output frequency is inversely proportional tocapacitor CF. CF values can be selected to producefrequencies above 20MHz.

A sine-shaping circuit converts the oscillator trianglewave into a low-distortion sine wave with constantamplitude. The triangle, square, and sine waves areinput to a multiplexer. Two address lines, A0 and A1,control which of the three waveforms is selected. Theoutput amplifier produces a constant 2VP-P amplitude(±1V), regardless of wave shape or frequency.

The triangle wave is also sent to a comparator that pro-duces a high-speed square-wave SYNC waveform thatcan be used to synchronize other oscillators. The SYNCcircuit has separate power-supply leads and can bedisabled.

Two other phase-quadrature square waves are gener-ated in the basic oscillator and sent to one side of an“exclusive-OR” phase detector. The other side of thephase-detector input (PDI) can be connected to anexternal oscillator. The phase-detector output (PDO) isa current source that can be connected directly toFADJ to synchronize the MAX038 with the externaloscillator.

Waveform SelectionThe MAX038 can produce either sine, square, or trian-gle waveforms. The TTL/CMOS-logic address pins (A0and A1) set the waveform, as shown below:

X = Don’t care.

Waveform switching can be done at any time, withoutregard to the phase of the output. Switching occurswithin 0.3µs, but there may be a small transient in theoutput waveform that lasts 0.5µs.

Waveform TimingOutput Frequency

The output frequency is determined by the currentinjected into the IIN pin, the COSC capacitance (toground), and the voltage on the FADJ pin. When

VFADJ = 0V, the fundamental output frequency (Fo) isgiven by the formula:

Fo (MHz) = IIN (µA) ÷ CF (pF) [1]

The period (to) is:

to (µs) = CF (pF) ÷ IIN (µA) [2]

where:

IIN = current injected into IIN (between 2µA and 750µA)

CF = capacitance connected to COSC and GND (20pF to >100µF).

For example:

0.5MHz = 100µA ÷ 200pF

and

2µs = 200pF ÷ 100µA

Optimum performance is achieved with IIN between10µA and 400µA, although linearity is good with IINbetween 2µA and 750µA. Current levels outside of thisrange are not recommended. For fixed-frequency oper-ation, set IIN to approximately 100µA and select a suit-able capacitor value. This current produces the lowesttemperature coefficient, and produces the lowest fre-quency shift when varying the duty cycle.

The capacitance can range from 20pF to more than100µF, but stray circuit capacitance must be minimizedby using short traces. Surround the COSC pin and thetrace leading to it with a ground plane to minimize cou-pling of extraneous signals to this node. Oscillationabove 20MHz is possible, but waveform distortionincreases under these conditions. The low frequencylimit is set by the leakage of the COSC capacitor andby the required accuracy of the output frequency.Lowest frequency operation with good accuracy is usu-ally achieved with 10µF or greater non-polarizedcapacitors.

An internal closed-loop amplifier forces IIN to virtualground, with an input offset voltage less than ±2mV. IINmay be driven with either a current source (IIN), or avoltage (VIN) in series with a resistor (RIN). (A resistorbetween REF and IIN provides a convenient method ofgenerating IIN: IIN = VREF/RIN.) When using a voltagein series with a resistor, the formula for the oscillator fre-quency is:

Fo (MHz) = VIN ÷ [RIN x CF (pF)] [3]

and:

to (µs) = CF (pF) x RIN ÷ VIN [4]

MA

X0

38

High-Frequency Waveform Generator

_______________________________________________________________________________________ 9

A0 A1 WAVEFORM

X 1 Sine wave0 0 Square wave

1 0 Triangle wave

MA

X0

38 When the MAX038’s frequency is controlled by a volt-

age source (VIN) in series with a fixed resistor (RIN), theoutput frequency is a direct function of VIN as shown inthe above equations. Varying VIN modulates the oscilla-tor frequency. For example, using a 10kΩ resistor forRIN and sweeping VIN from 20mV to 7.5V produceslarge frequency deviations (up to 375:1). Select RIN sothat IIN stays within the 2µA to 750µA range. The band-width of the IIN control amplifier, which limits the modu-lating signal’s highest frequency, is typically 2MHz.

IIN can be used as a summing point to add or subtractcurrents from several sources. This allows the outputfrequency to be a function of the sum of several vari-ables. As VIN approaches 0V, the IIN error increasesdue to the offset voltage of IIN.

Output frequency will be offset 1% from its final valuefor 10 seconds after power-up.

FADJ InputThe output frequency can be modulated by FADJ,which is intended principally for fine frequency control,usually inside phase-locked loops. Once the funda-mental, or center frequency (Fo) is set by IIN, it may bechanged further by setting FADJ to a voltage other than0V. This voltage can vary from -2.4V to +2.4V, causingthe output frequency to vary from 1.7 to 0.30 times thevalue when FADJ is 0V (Fo ±70%). Voltages beyond±2.4V can cause instability or cause the frequencychange to reverse slope.

The voltage on FADJ required to cause the output todeviate from Fo by Dx (expressed in %) is given by theformula:

VFADJ = -0.0343 x Dx [5]

where VFADJ, the voltage on FADJ, is between-2.4V and +2.4V.

Note: While IIN is directly proportional to the fundamen-tal, or center frequency (Fo), VFADJ is linearly related to% deviation from Fo. VFADJ goes to either side of 0V,corresponding to plus and minus deviation.

The voltage on FADJ for any frequency is given by theformula:

VFADJ = (Fo - Fx) ÷ (0.2915 x Fo) [6]

where:

Fx = output frequency

Fo = frequency when VFADJ = 0V.

Likewise, for period calculations:

VFADJ = 3.43 x (tx - to) ÷ tx [7]

where:

tx = output period

to = period when VFADJ = 0V.

Conversely, if VFADJ is known, the frequency is givenby:

Fx = Fo x (1 - [0.2915 x VFADJ]) [8]

and the period (tx) is:

tx = to ÷ (1 - [0.2915 x VFADJ]) [9]

Programming FADJ FADJ has a 250µA constant current sink to V- that mustbe furnished by the voltage source. The source is usu-ally an op-amp output, and the temperature coefficientof the current sink becomes unimportant. For manualadjustment of the deviation, a variable resistor can beused to set VFADJ, but then the 250µA current sink’stemperature coefficient becomes significant. Sinceexternal resistors cannot match the internal tempera-ture-coefficient curve, using external resistors to pro-gram VFADJ is intended only for manual operation,when the operator can correct for any errors. Thisrestriction does not apply when VFADJ is a true voltagesource.

A variable resistor, RF, connected between REF (+2.5V)and FADJ provides a convenient means of manuallysetting the frequency deviation. The resistance value(RF) is:

RF = (VREF - VFADJ) ÷ 250µA [10]

VREF and VFADJ are signed numbers, so use correctalgebraic convention. For example, if VFADJ is -2.0V(+58.3% deviation), the formula becomes:

RF = (+2.5V - (-2.0V)) ÷ 250µA

= (4.5V) ÷ 250µA

= 18kΩ

Disabling FADJ The FADJ circuit adds a small temperature coefficientto the output frequency. For critical open-loop applica-tions, it can be turned off by connecting FADJ to GND(not REF) through a 12kΩ resistor (R1 in Figure 2). The -250µA current sink at FADJ causes -3V to be devel-oped across this resistor, producing two results. First,the FADJ circuit remains in its linear region, but discon-nects itself from the main oscillator, improving tempera-ture stability. Second, the oscillator frequency doubles.If FADJ is turned off in this manner, be sure to correctequations 1-4 and 6-9 above, and 12 and 14 below bydoubling Fo or halving to. Although this method doublesthe normal output frequency, it does not double theupper frequency limit. Do not operate FADJ open cir-cuit or with voltages more negative than -3.5V. Doingso may cause transistor saturation inside the IC, lead-ing to unwanted changes in frequency and duty cycle.

High-Frequency Waveform Generator

10 ______________________________________________________________________________________

With FADJ disabled, the output frequency can still bechanged by modulating IIN.

Swept Frequency OperationThe output frequency can be swept by applying a vary-ing signal to IIN or FADJ. IIN has a wider range, slightlyslower response, lower temperature coefficient, andrequires a single polarity current source. FADJ may beused when the swept range is less than ±70% of thecenter frequency, and it is suitable for phase-lockedloops and other low-deviation, high-accuracy closed-loop controls. It uses a sweeping voltage symmetricalabout ground.

Connecting a resistive network between REF, the volt-age source, and FADJ or IIN is a convenient means ofoffsetting the sweep voltage.

Duty CycleThe voltage on DADJ controls the waveform duty cycle(defined as the percentage of time that the outputwaveform is positive). Normally, VDADJ = 0V, and theduty cycle is 50% (Figure 2). Varying this voltage from+2.3V to -2.3V causes the output duty cycle to varyfrom 15% to 85%, about -15% per volt. Voltagesbeyond ±2.3V can shift the output frequency and/orcause instability.

DADJ can be used to reduce the sine-wave distortion.The unadjusted duty cycle (VDADJ = 0V) is 50% ±2%;any deviation from exactly 50% causes even order har-monics to be generated. By applying a smalladjustable voltage (typically less than ±100mV) toVDADJ, exact symmetry can be attained and the distor-tion can be minimized (see Figure 2).

The voltage on DADJ needed to produce a specificduty cycle is given by the formula:

VDADJ = (50% - dc) x 0.0575 [11]

or:

VDADJ = (0.5 - [tON ÷ to]) x 5.75 [12]

where:

VDADJ = DADJ voltage (observe the polarity)

dc = duty cycle (in %)

tON = ON (positive) time

to = waveform period.

Conversely, if VDADJ is known, the duty cycle and ONtime are given by:

dc = 50% - (VDADJ x 17.4) [13]

tON = to x (0.5 - [VDADJ x 0.174]) [14]

MA

X0

38

High-Frequency Waveform Generator

______________________________________________________________________________________ 11

MAX038

1µF

GND

COSC12

AO

V-

1811926GND GNDGND GND

5

8

10

7

1

13

14

15

16 N.C.

3

FADJ

IIN

DADJ

REF

OUT

DV+

DGND

SYNC

PDI

PDO

V+ A141720

–5V +5V

C2

1nFC3

1µFC1

12kΩR1

20kΩRIN

FREQUENCY

50ΩR2

N.C.

CF

19 SINE-WAVEOUTPUT

2 x 2.5VRIN x CF

Fo =

MAX038

100kΩR5

5kΩ R6

100kΩR7

100kΩR3

100kΩR4

DADJ

REF

+2.5V–2.5V

PRECISION DUTY-CYCLE ADJUSTMENT CIRCUIT

ADJUST R6 FOR MINIMUM SINE-WAVE DISTORTION

Figure 2. Operating Circuit with Sine-Wave Output and 50% Duty Cycle; SYNC and FADJ Disabled

MA

X0

38 Programming DADJ

DADJ is similar to FADJ; it has a 250µA constant cur-rent sink to V- that must be furnished by the voltagesource. The source is usually an op-amp output, andthe temperature coefficient of the current sink becomesunimportant. For manual adjustment of the duty cycle, avariable resistor can be used to set VDADJ, but then the250µA current sink’s temperature coefficient becomessignificant. Since external resistors cannot match theinternal temperature-coefficient curve, using externalresistors to program VDADJ is intended only for manualoperation, when the operator can correct for any errors.This restriction does not apply when VDADJ is a truevoltage source.

A variable resistor, RD, connected between REF(+2.5V) and DADJ provides a convenient means ofmanually setting the duty cycle. The resistance value(RD) is:

RD = (VREF - VDADJ) ÷ 250µA [15]

Note that both VREF and VDADJ are signed values, soobserve correct algebraic convention. For example, ifVDADJ is -1.5V (23% duty cycle), the formula becomes:

RD = (+2.5V - (-1.5V)) ÷ 250µA

= (4.0V) ÷ 250µA = 16kΩVarying the duty cycle in the range 15% to 85% hasminimal effect on the output frequency—typically lessthan 2% when 25µA < IIN < 250µA. The DADJ circuit iswideband, and can be modulated at up to 2MHz (seephotos, Typical Operating Characteristics).

OutputThe output amplitude is fixed at 2VP-P, symmetricalaround ground, for all output waveforms. OUT has anoutput resistance of under 0.1Ω, and can drive ±20mAwith up to a 50pF load. Isolate higher output capaci-tance from OUT with a resistor (typically 50Ω) or bufferamplifier.

Reference VoltageREF is a stable 2.50V bandgap voltage reference capa-ble of sourcing 4mA or sinking 100µA. It is principallyused to furnish a stable current to IIN or to bias DADJand FADJ. It can also be used for other applicationsexternal to the MAX038. Bypass REF with 100nF to min-imize noise.

Selecting Resistors and CapacitorsThe MAX038 produces a stable output frequency overtime and temperature, but the capacitor and resistorsthat determine frequency can degrade performance ifthey are not carefully chosen. Resistors should bemetal film, 1% or better. Capacitors should be chosen

for low temperature coefficient over the whole tempera-ture range. NPO ceramics are usually satisfactory.

The voltage on COSC is a triangle wave that variesbetween 0V and -1V. Polarized capacitors are generallynot recommended (because of their outrageous tem-perature dependence and leakage currents), but if theyare used, the negative terminal should be connected toCOSC and the positive terminal to GND. Large-valuecapacitors, necessary for very low frequencies, shouldbe chosen with care, since potentially large leakagecurrents and high dielectric absorption can interferewith the orderly charge and discharge of CF. If possi-ble, for a given frequency, use lower IIN currents toreduce the size of the capacitor.

SYNC OutputSYNC is a TTL/CMOS-compatible output that can beused to synchronize external circuits. The SYNC outputis a square wave whose rising edge coincides with theoutput rising sine or triangle wave as it crosses through0V. When the square wave is selected, the rising edgeof SYNC occurs in the middle of the positive half of theoutput square wave, effectively 90° ahead of the output.The SYNC duty cycle is fixed at 50% and is indepen-dent of the DADJ control.

Because SYNC is a very-high-speed TTL output, thehigh-speed transient currents in DGND and DV+ canradiate energy into the output circuit, causing a narrowspike in the output waveform. (This spike is difficult tosee with oscilloscopes having less than 100MHz band-width). The inductance and capacitance of IC socketstend to amplify this effect, so sockets are not recom-mended when SYNC is on. SYNC is powered from sep-arate ground and supply pins (DGND and DV+), and itcan be turned off by making DV+ open circuit. If syn-chronization of external circuits is not used, turning offSYNC by DV+ opening eliminates the spike.

Phase DetectorsInternal Phase Detector

The MAX038 contains a TTL/CMOS phase detector thatcan be used in a phase-locked loop (PLL) to synchro-nize its output to an external signal (Figure 3). Theexternal source is connected to the phase-detectorinput (PDI) and the phase-detector output is taken fromPDO. PDO is the output of an exclusive-OR gate, andproduces a rectangular current waveform at theMAX038 output frequency, even with PDI grounded.PDO is normally connected to FADJ and a resistor,RPD, and a capacitor CPD, to GND. RPD sets the gainof the phase detector, while the capacitor attenuateshigh-frequency components and forms a pole in thephase-locked loop filter.

High-Frequency Waveform Generator

12 ______________________________________________________________________________________

PDO is a rectangular current-pulse train, alternatingbetween 0µA and 500µA. It has a 50% duty cycle whenthe MAX038 output and PDI are in phase-quadrature(90° out of phase). The duty cycle approaches 100%as the phase difference approaches 180° and con-versely, approaches 0% as the phase differenceapproaches 0°. The gain of the phase detector (KD)can be expressed as:

KD = 0.318 x RPD (volts/radian) [16]

where RPD = phase-detector gain-setting resistor.

When the loop is in lock, the input signals to the phasedetector are in approximate phase quadrature, the dutycycle is 50%, and the average current at PDO is 250µA(the current sink of FADJ). This current is dividedbetween FADJ and RPD; 250µA always goes into FADJand any difference current is developed across RPD,creating VFADJ (both polarities). For example, as thephase difference increases, PDO duty cycle increases,the average current increases, and the voltage on RPD(and VFADJ) becomes more positive. This in turndecreases the oscillator frequency, reducing the phasedifference, thus maintaining phase lock. The higherRPD is, the greater VFADJ is for a given phase differ-ence; in other words, the greater the loop gain, the lessthe capture range. The current from PDO must also

charge CPD, so the rate at which VFADJ changes (theloop bandwidth) is inversely proportional to CPD.

The phase error (deviation from phase quadrature)depends on the open-loop gain of the PLL and the ini-tial frequency deviation of the oscillator from the exter-nal signal source. The oscillator conversion gain (Ko) is:

KO = ∆ωo ÷ ∆VFADJ [17]

which, from equation [6] is:

KO = 3.43 x ωo (radians/sec) [18]

The loop gain of the PLL system (KV) is:

KV = KD x KO [19]

where:

KD = detector gain

KO = oscillator gain.

With a loop filter having a response F(s), the open-looptransfer function, T(s), is:

T(s) = KD x KO x F(s) ÷ s [20]

Using linear feedback analysis techniques, the closed-loop transfer characteristic, H(s), can be related to theopen-loop transfer function as follows:

H(s) = T(s) ÷ [1+ T(s)] [21]

The transient performance and the frequency responseof the PLL depends on the choice of the filter charac-teristic, F(s).

When the MAX038 internal phase detector is not used,PDI and PDO should be connected to GND.

External Phase DetectorsExternal phase detectors may be used instead of theinternal phase detector. The external phase detectorshown in Figure 4 duplicates the action of the MAX038’sinternal phase detector, but the optional ÷N circuit canbe placed between the SYNC output and the phasedetector in applications requiring synchronizing to anexact multiple of the external oscillator. The resistor net-work consisting of R4, R5, and R6 sets the sync range,while capacitor C4 sets the capture range. Note thatthis type of phase detector (with or without the ÷N cir-cuit) locks onto harmonics of the external oscillator aswell as the fundamental. With no external oscillatorinput, this circuit can be unpredictable, depending onthe state of the external input DC level.

Figure 4 shows a frequency phase detector that locksonto only the fundamental of the external oscillator.With no external oscillator input, the output of the fre-quency phase detector is a positive DC voltage, andthe oscillations are at the lowest frequency as set byR4, R5, and R6.

MA

X0

38

High-Frequency Waveform Generator

______________________________________________________________________________________ 13

Figure 3. Phase-Locked Loop Using Internal Phase Detector

MAX038

GND

COSC12

A0V-

181192 6GND GND

15DGNDGND GND

5

8

10

7

1

13

3

FADJ

IIN

DADJ

REF

RD

OUT

PDI

PDO

V+

17

DV+

16 20

+5V -5V

C21µF

C11µF

CENTERFREQUENCY

50ΩROUT

CF

RPD

CPD

19

RFOUTPUT

A14

SYNC

14

EXTERNAL OSC INPUT

MA

X0

38

High-Frequency Waveform Generator

14 ______________________________________________________________________________________

Figure 4. Phase-Locked Loop Using External Phase Detector

Figure 5. Phase-Locked Loop Using External Frequency Phase Detector

MAX038

GND

COSC12

A0V-

181192 6GND GND

15DGNDGND GND

5

8

10

7

1

13

3

FADJ

IIN

DADJ

REF

R2CW

R3

OUT

PDI

PDO

V+

17

DV+

16 20

+5V -5V

-5V

C21µF

C11µF

CENTERFREQUENCY

50ΩR1

R6GAIN

R5OFFSET

R4PHASE DETECTOR

EXTERNALOSC INPUT

C4CAPTURE

19 RFOUTPUT

A14

SYNC

14÷N

C3FREQUENCY

MAX038

GND

COSC12

A0V-

181192 6GND GND

15DGNDGND GND

5

8

10

7

1

13

3

FADJ

IIN

DADJ

REF

R2CW

R3

OUT

PDI

PDO

V+

17

DV+

16 20

+5V -5V

-5V

C21µF

C11µF

CENTERFREQUENCY

50ΩR1

R6GAIN

R5OFFSET

R4

C4CAPTURE

19

RFOUTPUT

A14

SYNC

14÷N

C3FREQUENCY

EXTERNALOSC INPUT

FREQUENCY PHASE DETECTOR

MA

X0

38

High-Frequency Waveform Generator

______________________________________________________________________________________ 15

N4N3 N2

MC1

4515

1

N6

8.19

2MHz

MAX

427

N5

OUT1

OUT2

RFB

VREF

VDD

GND1

MX7541

N7 N8 N9 T/R

N12

N13

N10

N11

OSC O

UT

OSC I

N

LDNN1 N0 FV PD

VPD

RRA

2RA

1

RA0

PD1 O

UT V DD

V SS

F IN

35pF

20pF

1514

281

GND

BIT1

BIT2

BIT3

BIT4

BIT5

BIT6

BIT12

BIT11

BIT10

BIT9

BIT8

BIT7

MAX

038

A0 A1 COSC

GND1

DADJ

FADJ

OUT

GND V+ DV+

DGND

SYNC PD

I

PDO

VREF

V-

GND1

IINGN

D1

3.3M

ΩPD

V

PDR

3.3M

Ω33

k Ω 0.1µ

F

0.1µ

F

33k Ω

0.

1µF

0.1µ

F

7.5k

Ω

10kΩ

2 3

7 4

6

0.1µ

F0.

1µF+2

.5V

±2.5

V

35 pF

1011

0.1µ

F0.

1µF

0.1µ

F 50

.0Ω

100Ω

120

50Ω

, 50M

HzLO

WPA

SS F

ILTE

R22

0nH

220n

H

56pF

110p

F56

pF

50Ω

SIGN

ALOU

TPUT

SYNC

OUTP

UT

+5V

-5V

9 10

1 18

3 2

1

0V T

O 2.

5V

2N39

04

3.33

k Ω

2.7M

1k Ω

1kΩ

568 4

72N

39061N

914

2µA

to75

0µA

MAX

412

MAX

412

8.192MHz4.096MHz2.048MHz1.024MHz

512kHz256kHz128kHz64kHz32kHz16kHz8kHz4kHz2kHz1kHz

WAV

EFOR

MSE

LECT

FREQ

UENC

Y SY

NTHE

SIZE

R 1k

Hz R

ESOL

UTIO

N; 8

kHz T

O 16

.383

MHz

Figure 6. Crystal-Controlled, Digitally Programmed Frequency Synthesizer—8kHz to 16MHz with 1kHz Resolution

MA

X0

38

High-Frequency Waveform Generator

Layout ConsiderationsRealizing the full performance of the MAX038 requirescareful attention to power-supply bypassing and boardlayout. Use a low-impedance ground plane, and con-nect all five GND pins directly to it. Bypass V+ and V-directly to the ground plane with 1µF ceramic capaci-tors or 1µF tantalum capacitors in parallel with 1nFceramics. Keep capacitor leads short (especially withthe 1nF ceramics) to minimize series inductance.

If SYNC is used, DV+ must be connected to V+, DGNDmust be connected to the ground plane, and a second1nF ceramic should be connected as close as possiblebetween DV+ and DGND (pins 16 and 15). It is notnecessary to use a separate supply or run separatetraces to DV+. If SYNC is disabled, leave DV+ open.Do not open DGND.

Minimize the trace area around COSC (and the groundplane area under COSC) to reduce parasitic capaci-tance, and surround this trace with ground to preventcoupling with other signals. Take similar precautionswith DADJ, FADJ, and IIN. Place CF so its connectionto the ground plane is close to pin 6 (GND).

___________Applications InformationFrequency Synthesizer

Figure 6 shows a frequency synthesizer that producesaccurate and stable sine, square, or triangle waves witha frequency range of 8kHz to 16.383MHz in 1kHz incre-ments. A Motorola MC145151 provides the crystal-con-trolled oscillator, the ÷N circuit, and a high-speed phasedetector. The manual switches set the output frequency;opening any switch increases the output frequency.Each switch controls both the ÷N output and anMX7541 12-bit DAC, whose output is converted to a cur-rent by using both halves of the MAX412 op amp. Thiscurrent goes to the MAX038 IIN pin, setting its coarsefrequency over a very wide range.

Fine frequency control (and phase lock) is achievedfrom the MC145151 phase detector through the differ-ential amplifier and lowpass filter, U5. The phase detec-

tor compares the ÷N output with the MAX038 SYNCoutput and sends differential phase information to U5.U5’s single-ended output is summed with an offset intothe FADJ input. (Using the DAC and the IIN pin forcoarse frequency control allows the FADJ pin to havevery fine control with reasonably fast response to switchchanges.)

A 50MHz, 50Ω lowpass filter in the output allows pas-sage of 16MHz square waves and triangle waves withreasonable fidelity, while stopping high-frequency noisegenerated by the ÷N circuit.

V+

PDI

SYNC

AO

DADJ

PDOFADJ

0.118"(2.997mm)

0.106"(2.692mm)

A1

COSC

GND

IINGND GND

DGND

DV+

GND

GND REF V- OUT

TRANSISTOR COUNT: 855SUBSTRATE CONNECTED TO GND

___________________Chip Topography

Maxim cannot assume responsibility for use of any circuitry other than circuitry entirely embodied in a Maxim product. No circuit patent licenses areimplied. Maxim reserves the right to change the circuitry and specifications without notice at any time.

16 ____________________Maxim Integrated Products, 120 San Gabriel Drive, Sunnyvale, CA 94086 408-737-7600

© 2004 Maxim Integrated Products Printed USA is a registered trademark of Maxim Integrated Products.

Package InformationFor the latest package outline information, go towww.maxim-ic.com/packages.

This datasheet has been download from:

www.datasheetcatalog.com

Datasheets for electronics components.

DEIC51515 Ampere Low-Side Ultrafast RF MOSFET Driver

Features • Built using the advantages and compatibility of CMOS and IXYS HDMOSTM processes • Latch-Up Protected • High Peak Output Current: 15A Peak • Wide Operating Range: 8V to 30V • Rise And Fall Times of <4ns • Minimum Pulse Width Of 8ns • High Capacitive Load Drive Capability: 2nF in <4ns • Matched Rise And Fall Times • 18ns Input To Output Delay Time • Low Output Impedance • Low Quiescent Supply Currentt Applications • Driving RF MOSFETs • Class D or E Switching Amplifier Drivers • Multi MHz Switch Mode Power Supplies (SMPS) • Pulse Generators • Acoustic Transducer Drivers • Pulsed Laser Diode Drivers • DC to DC Converters • Pulse Transformer Driver

Description TheDEIC515 is a CMOS high speed high current gate driver specifically designed to drive MOSFETs in Class D and E HF RF applications at up to 45MHz, as well as other applications requiring ultrafast rise and fall times or short minimum pulse widths. The DEIC515 can source and sink 15A of peak current while producing voltage rise and fall times of less than 4ns, and minimum pulse widths of 8ns. The input of the driver is fully immune to latch up over the entire operating range. Designed with small internal delays, cross conduction/current shoot-through is virtually eliminated in the DEIC515. Its features and wide safety margin in operating voltage and power make the DEIC515 unmatched in performance and value.

The DEIC515 is packaged in DEI's low inductance RF package incorporating DEI's patented (1) RF layout techniques to minimize stray lead inductances for optimum switching performance. The DEIC515 is a surface-mount device. (1) DEI U.S. Patent #4,891,686

Figure 1 - DEIC515 Functional Diagram

IN GND

IN

VCC IN

DGND

OUT

VCC

DEIC51515 Ampere Low-Side Ultrafast RF MOSFET Driver

Absolute Maximum Ratings

Parameter ValueMaximum Junction Temperature 150oC Operating Temperature Range -40oC to 85oC Thermal Impedance (Junction To Case) θJC 0.13oC/W

Electrical Characteristics Unless otherwise noted, TA = 25 oC, 8V < VCC =VCCIN < 30V . All voltage measurements with respect to DGND. DEIC515 configured as described in Test Conditions.

Symbol Parameter Test Conditions Min Typ Max Units

VIH High input voltage VCCIN -2 V VIL Low input voltage 0.8 V VIN Input voltage range -5 VCC + 0.3 V IIN Input current 0V≤ VIN ≤VCC,VCCIN -10 10 µA VOH High output voltage VCC,VCCIN - .025 V VOL Low output voltage 0.025 V ROH Output resistance

@ Output high IOUT = 10mA, VCC = 15V 0.55 0.85 Ω

ROL Output resistance @ Output Low

IOUT = 10mA, VCC = 15V 0.35 0.85 Ω

IPEAK Peak output current VCC,VCCIN = 15V 15 A IDC Continuous output

current 2.5 A

fMAX Maximum frequency CL=2nF VCC,VCCIN =15V 45 MHz tR Rise time (1) CL=1nF VCC,VCCIN =15V VOH=2V to 12V

CL=2nF VCC,VCCIN =15V VOH=2V to 12V 2.5

4.1 ns

ns tF Fall time (1) CL=1nF VCC,VCCIN =15V VOH=12V to 2V

CL=2nF VCC,VCCIN =15V VOH=12V to 2V 2.5

3.9 ns

ns tONDLY On-time propagation

delay (1) CL=2nF Vcc=15V 17.4 18.5 ns

tOFFDLY Off-time propagation delay (1)

CL=2nF Vcc=15V 14.6 16 ns

PWmin Minimum pulse width FWHM CL=1nF VCC,VCCIN =15V +3V to +3V CL=1nF VCC,VCCIN =15V

6.4 8.2

ns ns

VCC,VCCIN Power supply voltage 8 15 30 V ICC Power supply current VIN = 0V

VIN = VCCIN 0 10

10 µA µA

Parameter Value

Supply Voltage 30V

All other Pins -0.3V to (Vcc,Vccin)+0.3V

Power Dissipation TAMBIENT≤25C Tcase≤25C

2W 100W

Storage Temperature -40C to 150C

Soldering Lead Temperature (10 seconds maximum)

300C

Input -5V to Vccin+0.3V

DEIC51515 Ampere Low-Side Ultrafast RF MOSFET Driver

Lead Description - DEIC515

Note 1: Operating the device beyond parameters with listed “absolute maximum ratings” may cause permanent damage to the device. Typical values indicate conditions for which the device is intended to be functional, but do not guarantee specific performance limits. The guaranteed specifications apply only for the test conditions listed. Exposure

CAUTION: These devices are sensitive to electrostatic discharge; follow proper ESD procedures when handling and assembling this component.

Figure 2 - DEIC515 Package Photo And Outline

SYMBOL FUNCTION DESCRIPTION

VCC Output Supply Voltage Input for the positive output section power-supply voltage. These leads provide power to the output section. Both leads must be connected.

IN Input Input signal. OUT Output Driver Output.

PGND Power Ground The system ground leads. Internally connected to all circuitry, these leads provide ground reference for the entire chip. These leads should be connected to a low noise analog ground plane for optimum performance.

INGND Input Ground The input ground lead. This lead is a Kelvin connection internally to PGND. This lead must not be connected to PGND as excessive current can damage this lead.

VCCIN Supply Voltage Input for the positive input section power-supply voltage. This lead provide power to the input section. This lead should not be directly connected to VCC.

Bottom View

DEIC51515 Ampere Low-Side Ultrafast RF MOSFET Driver

Typical Performance Characteristics

Figure 3a - Characteristics Test Diagram

Application The very high currents and high speeds inside the DEIC515 create very large transients. To avoid problems with false triggering, the input to the DEIC515 should be supplied via a common mode choke. This is a simple tri-filar winding on a small ferrite core. This prevents high speed transients from effecting the input signals, by allowing the input signals to follow the internal die potential changes without changing the state of the input.

++

--

Input

INVCC VCC IN OUT INGND GND

CL

VCC

L1

10µF 10µF

DE275X2-102N06A

RF Power MOSFET

VDSS = 1000 V

ID25 = 16 A

RDS(on) = 0.8 ΩΩΩΩ

PDC = 1180 W

Symbol Test Conditions Maximum Ratings

VDSS TJ = 25°C to 150°C 1000 V

VDGR TJ = 25°C to 150°C; RGS = 1 MΩ 1000 V

VGS Continuous ±20 V

VGSM Transient ±30 V

ID25 Tc = 25°C 16 A

IDM Tc = 25°C, pulse width limited by TJM 48 A

IAR Tc = 25°C 6 A

EAR Tc = 25°C 20 mJ

dv/dt

IS ≤ IDM, di/dt ≤ 100A/µs, VDD ≤ VDSS, Tj ≤ 150°C, RG = 0.2Ω

5 V/ns

IS = 0 >200 V/ns

PDC (1) 1180 W

PDHS (1) Tc = 25°C, Derate 5.0W/°C above 25°C 750 W

PDAMB (1) Tc = 25°C 5.0 W

Symbol Test Conditions Characteristic Values TJ = 25°C unless otherwise specified

min. typ. max.

VDSS VGS = 0 V, ID = 3 ma 1000 V

VGS(th) VDS = VGS, ID = 4 ma 2.5 5.5 V

IGSS VGS = ±20 VDC, VDS = 0 ±100 nA

IDSS VDS = 0.8 VDSS TJ = 25°C VGS = 0 TJ = 125°C

50 1

µA mA

RDS(on) 1.6 Ω

gfs VDS = 15 V, ID = 0.5ID25, pulse test 2 7.5 S

VGS = 15 V, ID = 0.5ID25 Pulse test, t ≤ 300µS, duty cycle d ≤ 2%

RthJC (1) 0.25 C/W

RthJHS (1) 0.50 C/W

TJ -55 +175 °C

TJM 175 °C

Tstg -55 +175 °C

TL 1.6mm (0.063 in) from case for 10 s 300 °C

Weight 4 g

Features

• Isolated Substrate

− high isolation voltage (>2500V)

− excellent thermal transfer

− Increased temperature and power cycling capability

• IXYS advanced low Qg process

• Low gate charge and capacitances

− easier to drive

− faster switching

• Low RDS(on)

• Very low insertion inductance (<2nH)

• No beryllium oxide (BeO) or other hazardous materials

Advantages

• High Performance Push-Pull RF Package

• Optimized for RF and high speed switching at frequencies to >100MHz

• Easy to mount—no insulators needed

• High power density

♦ Common Source Push-Pull Pair ♦ N-Channel Enhancement Mode ♦ Low Qg and Rg ♦ High dv/dt ♦ Nanosecond Switching

DRAIN 1

SG1 SD1

GATE 1

DRAIN 2

SG2SD2

GATE 2

The DE275X2-102N06A is a matched pair of RF power MOSFET devices in a common source configuration. The device is optimized for push-pull or paral-lel operation in RF generators and amplifiers at frequencies to >65 MHz.

Note: All specifications are per each transistor, unless otherwise noted. (1) Thermal specifications are for the package, not per transistor

Unless noted, specifications are for each output device

Source 1 Source 2

DE275X2-102N06A

RF Power MOSFET

Symbol Test Conditions Characteristic Values (TJ = 25°C unless otherwise specified)

min. typ. max.

RG 0.3 Ω

Ciss 1800 pF

Coss VGS = 0 V, VDS = 0.8 VDSS(max), f = 1 MHz

130 pF

Crss 25 pF

Td(on) 3 ns

Ton VGS = 15 V, VDS = 0.8 VDSS

ID = 0.5 IDM RG = 0.2 Ω (External)

2 ns

Td(off) 4 ns

Toff 5 ns

Qg(on) 50 nC

Qgs VGS = 10 V, VDS = 0.5 VDSS

ID = 0.5 ID25 20 nC

Qgd 30 nC

Cstray Back Metal to any Pin 21 pF

Source-Drain Diode Characteristic Values (TJ = 25°C unless otherwise specified)

Symbol Test Conditions min. typ. max.

IS VGS = 0 V 6 A

ISM Repetitive; pulse width limited by TJM 96 A

VSD 1.5 V

Trr 200 ns

IF = IS, VGS = 0 V, Pulse test, t ≤ 300 µs, duty cycle ≤ 2%

QRM IF = IS, -di/dt = 100A/µs, VR = 100V

0.6 µC

IRM 4 A

IXYS RF reserves the right to change limits, test conditions and dimensions.

IXYS RF MOSFETS are covered by one or more of the following U.S. patents:

4,835,592 4,860,072 4,881,106 4,891,686 4,931,844 5,017,508

5,034,796 5,049,961 5,063,307 5,187,117 5,237,481 5,486,715

5,381,025 5,640,045

(1) These parameters apply to the package, not individual MOSFET devices. For detailed device mounting and installation instructions, see the “DE-Series MOSFET Mounting Instructions” technical note on IXYS RF’s web site at www.ixysrf.com/Technical_Support/App_notes.html

DE275X2-102N06A

RF Power MOSFET

10

100

1000

10000

0 100 200 300 400 500 600 700 800 900 1000

Vds in Volts

Capacitance in pF

Ciss

Coss

Crss

275X2-102N06A Capacitances vs Vds

S = S1 = Source1 S = S1 = Source1

S = S2 = Source2 S = S2 = Source2

G1 = Gate1

G2 = Gate2

D1 = Drain1

D2 = Drain2

Note: Sources S1, S2 are independent, having no com-mon connection between them for the package diagram.

DE275X2-102N06A

RF Power MOSFET

102N06A DE-SERIES SPICE Model

The DE-SERIES SPICE Model is illustrated in Figure 1. The model is an expansion of the SPICE level 3 MOSFET model. It includes the stray inductive terms LG, LS and LD. Rd is the RDS(ON) of the device, Rds is the resistive leakage term. The output capacitance, COSS, and reverse transfer capacitance, CRSS are modeled with reversed biased diodes. This provides a varactor type response necessary for a high power device model. The turn on delay and the turn off delay are adjusted via Ron and Roff.

Figure 1 DE-SERIES SPICE Model

This SPICE model may be downloaded as a text file from the IXYS RF web site at www.ixysrf.com

Net List: *SYM=POWMOSN .SUBCKT 102N06A 10 20 30 * TERMINALS: D G S * 1000 Volt 6 Amp 1.6 Ohm N-Channel Power MOSFET M1 1 2 3 3 DMOS L=1U W=1U RON 5 6 .5 DON 6 2 D1 ROF 5 7 1.0 DOF 2 7 D1 D1CRS 2 8 D2 D2CRS 1 8 D2 CGS 2 3 1.9N RD 4 1 1.6 DCOS 3 1 D3 RDS 1 3 5.0MEG LS 3 30 .5N LD 10 4 1N LG 20 5 1N .MODEL DMOS NMOS (LEVEL=3 VTO=4 KP=2.3) .MODEL D1 D (IS=.5F CJO=10P BV=100 M=.5 VJ=.2 TT=1N) .MODEL D2 D (IS=.5F CJO=400P BV=1000 M=.6 VJ=.6 TT=1N RS=10M) .MODEL D3 D (IS=.5F CJO=400P BV=1000 M=.35 VJ=.6 TT=400N RS=10M) .ENDS

5 6

7

8

4

10 DRAIN

30 SOURCE

20 GATE

Don

Dcos

D2crs

D1crs

Rds

Ron

Doff

RoffRd

Lg

Ld

Ls

M32

13

An IXYS Company 2401 Research Blvd., Suite 108 Fort Collins, CO USA 80526 970-493-1901 Fax: 970-493-1903 Email: deiinfo@directedenergy.com Web: http://www.directedenergy.com

Doc #9200-0224 Rev 6 © 2006 IXYS RF