finalul

Transylvania University of Brasov Degree Project Faculty of Electrical Engineering and Computer Science 2011 Study Program: Electrical Engineering and Computers

3

C O N T E N T S

Abstract (RO) 1

Abstract (EN) 2

Objectives 3

Disposition 4

CHAPTER 1 – Theoretical Analysis of the EMI, PWM and Shielding

1.1 Introduction to EMI 5

1.2 Introduction to PWM 8

1.3 Shielding Effectiveness 11

CHAPTER 2 – Platform Design

2.1 Block Diagram 12

2.2 Circuit Schematic 13

2.3 General Design 14

2.4 The IC 16

2.5 GDT 18

2.6 GDT Waveform Correction Table 19

2.7 GDT Calculus 24

2.8 Short Guide for Designing a GDT 25

2.9 Flyback Transformer 26

2.10 Unprotected Circuit Waveforms 27

2.11 Circuit Parts 37


4

2.12 PCB Construction 38

2.13 Faraday’s Cage 38

2.14 MOSFETs 39

2.15 Physical Construction 41

CHAPTER 3 – Optimization of the Circuit by Different Means of Protection

3.1 Cable Shielding 45

3.2 Transformer Shielding 45

3.3 Shorten Paths 45

3.4 Hazards 46

CHAPTER 4 – Laboratory Project

4.1 Laboratory Project Proposal 48

Conclusion 49

Bibliography 50

Appendix

1. List of Used Abbreviations

2. SG3525 datasheet

3. IRFP250 datasheet


5

Abstract

Scopul acestei lucrări este de a arăta că prin aplicarea legilor de bază ale fizicii, arcul

electric considerat ca o “întrerupere” în circuit ce îşi crează propriul spaţiu conductor după

amorsarea sa, acţionează ca o sursă de perturbaţie preponderant radiantă. Ca urmare a fost

proiectat şi realizat un sistem funcţional, stabil şi repetabil ce poate fi utilizat la studiul

interferenţelor electromagnetice dintre acesta şi o gamă largă de circuite protejate sau

neprotejate.

Lucrărea de diplomă este de o importanţă majoră datorită informaţiilor teoretice

sintetizate şi prezentate, care ajută un inginer în devenire să înţeleagă riscurile întâmpinate la

construcţia unui astfel de circuit, principiile de funcţionare ale arcurilor electrice,

tranzistorilor cu efect de câmp de putere şi a transformatoarelor de izolare galvanică şi

comandă a tranzistorilor, precum şi cuplajele parasite ce pot influenţa funcţionarea circuitelor

proiectate de acesta.

Explicând şi folosind procesele fizice - ionizarea şi încălzirea termică în combinaţie

cu modularea în lăţime a pulsurilor PWM, modificând comanda tranzistorilor putem obţine

prin intermediul sursei de perturbaţie studiată (arcul electric în cazul acesta) un sunet

omnidirectional foarte clar reprodus, introdus de orice sursă de semnal audio.

Sistemul proiectat este funcţional şi reprezintă rezultatul unei munci depuse având

fonduri reduse şi pe o perioadă limitată de timp. Procesul de corecţie, design şi realizare a

circuitului nu a decurs întotdeauna fără piedici. O sumă de probleme au fost descoperite în

partea de verificare a sistemului şi cele mai importante sunt prezentate în această lucrare.

Circuitul studiat este conceput ca o platformă de laborator ce va folosi viitorilor

studenţi în analizele de cuplaje parazite introduse de prezenţa arcului electric asupra

circuitelor digitale.


6

Abstract

The result of this work is a functional and stable repeatable system. The system can be

used for studying the EMI on a wide range of shielded or unshielded circuits.

The importance of the project is highly ranked due to the theoretical information

presented in here that allows a prospective engineer to understand the hazards encountered

when building a circuit of this type and also the principles of operation of the HV arcs, power

FET transistors and GDT’s.

The object of this paper is to show that by applying the ordinary laws physics to the

arc considered as a gap in a circuit providing its own conductor by the volatilization of its

own material, all its principal phenomena can be accounted for, without the aid of a large

back EMF or of a “negative resistance” or of any other unusual attribute.

Also by combining physical processes like ionization and thermal heating in mixing

with a PWM circuit, modifying MOSFET’s command we can obtain a very clear reproduced

sound coming from a wide range of audio players.

The functional system is the result of hard work. The progress of the design work for

the creation of the arc did not always go smooth. A lot of problems were discovered in the

system verification process, and the most important are reported in this paper.

However, in the end, a functional system appeared.


7

Objectives:

- Obtaining a variable frequency electric arc of ~3 cm in length (2.8 cm) acting as a

perturbation source on its own circuit

- Modify the arc frequency of operation and study the waveforms of the key signals

driving the circuit

- Gate Drive Transformer correction

- Cancelling the perturbations by using different methods and electronic devices

- The physical construction of the PWM circuit

- Obtaining modulated audio into the perturbation device (electric low frequency arc

~37.5 kHz)

- Constructing and stabilizing a laboratory platform with the in-here circuit

- What GDT is and why is necessary

- How sensible are changes in its performances at high frequencies related to its design

- Synthesize the information in order to be useful for a prospective engineer, providing

guidance in understanding the studied phenomena


8

Disposition

This paper consists of three parts: one theoretical part about EMI, PWM and shielding,

one part about design and verification of a test platform for effectively seeing the corrected

waves in the circuit and some optimization methods.

The theoretical part consists of an explanation of EMI, PWM and shielding. It also

contains some useful formulas for theoretical calculation of some components used in the

circuit. The theoretical part guides the practical section and also providing extra theoretical

explanations regarding the studied component or block.

The design and verification part constitute a discussion of the thoughts initiating the

design and the proceedings leading to the completed test platform. The test system is used for

effectively seeing the principal waves present in the circuit.


9

CHAPTER 1 – Theoretical Analysis of the EMI, PWM and Shielding

1.1 Introduction to EMI

EMI (electromagnetic interference) is the disruption of operation of an electronic

device when it is in the vicinity of an electromagnetic field (EM field) in the radio frequency

(RF) spectrum that is caused by another electronic device.

The internal circuits of personal computers generate EM fields in the RF range. Also,

cathode ray tube (CRT) displays generate EM energy over a wide band of frequencies. These

emissions can interfere with the performance of sensitive wireless receivers nearby. Moderate

or high-powered wireless transmitters can produce EM fields strong enough to upset the

operation of electronic equipment nearby. If you live near a broadcast station or in the

downtown area of a large city, you have probably experienced EMI from radio or television

transmitters. Cordless telephones, home entertainment systems, computers, and certain

medical devices can fail to work properly in the presence of strong RF fields.

Problems with EMI can be minimized by ensuring that all electronic equipment is

operated with a good electrical ground system. In addition, cords and cables connecting the

peripherals in an electronic or computer system should, if possible, be shielded to keep

unwanted RF energy from entering or leaving. Specialized components such as line filters,

capacitors, and inductors can be installed in power cords and interconnecting cables to reduce

the EMI susceptibility of some systems. This is especially important if modifications might

void an existing warranty, and it is imperative with medical devices of any kind. [1]

Radio frequency (RF) is a rate of oscillation in the range of about 3 kHz to 300 GHz.


10

Types of Electromagnetic Interference (EMI)

Radiated Electromagnetic interference (EMI) or radio frequency interference (RFI) is

two types, Narrowband interference and Broadband interference. Narrowband EMI

interference usually arises from intentional transmissions such as radio and TV stations, pager

transmitters, cell phones, etc. Broadband EMI interference usually comes from incidental

radio frequency emitters. These include electric power transmission lines, electric motors,

thermostats, bug zappers, etc. [2]

Broadband EMI noise is stronger at low frequencies and diminishing at higher

frequencies, though this noise is often modulated, or varied, by the creating device in some

way. Broadband EMI/RFI noise is very difficult to filter it effectively once it has entered the

receiver chain.

Conducted electromagnetic interference is caused by the physical contact of the

conductors as opposed to radiated EMI which is caused by induction (without physical

contact of the conductors).

Electromagnetic Compatibility (EMC)

If there is no effect of the transmitters on the receivers it is called ‘electromagnetic

compatibility’, in short EMC.

Electromagnetic Compatibility (EMC) is defined as the ability of an equipment or

system to function satisfactorily in its electromagnetic environment without introducing

intolerable electromagnetic disturbances to anything in that environment. [3]

EMI can be intentionally used for radio jamming, as in some forms of electronic

warfare, or can occur unintentionally, as a result of spurious emissions for example through

intermodulation products, and the like. It frequently affects the reception of AM radio in

urban areas. It can also affect cell phone, FM radio and television reception, although to a

lesser extent. [4]


11

Low-Frequency EMF (electromagnetic field)

Common sources of low frequency EMF are the power supply of industry, households

and railways. Major sources of low-frequency EMF exposure include distribution and use of

electric power and transportation systems.

Receivers cannot only be radio receivers but every electronic device like computers,

measurement devices, control units, pacemakers etc.

High-Frequency EMF

The major sources for HF EMF are telecommunication facilities and associated

devices such as mobile telephones, medical, commercial and industrial equipment, radars, and

radio and television broadcast antennas.

Ionizing radiation

The range of the ionizing radiation starts at wavelength shorter than visible light,

meaning UV-light (380 nm) and includes X-ray, alpha and gamma radiation. Ionizing means

that the radiation has enough energy to free electrons from their atoms or molecules (UV

light) or even to change the structure of the atomic nucleus as technical used in nuclear power

stations. [5]

Maximum acceptable levels of EMI from electronic devices are detailed i.e. by the

FCC = Federal Communications Commission


12

1.2 Introduction to PWM

PWM or Pulse Width Modulation refers to the concept of rapidly pulsing the digital

signal of a wire to simulate a varying voltage on the wire. This method is commonly used for

driving motors, heaters, or lights in varying intensities or speeds.

A few terms are associated with PWM:

- Period - how long each complete pulse cycle takes

- Frequency - how often the pulses are generated. This value is typically specified in

Hz (cycles per second).

- Duty Cycle - refers to the amount of time in the period that the pulse is active or high.

Duty Cycle is typically specified as a percentage of the full period. [6]

Figure (1.1) [6] – Different duty cycles of PWM


13

Figure (1.2) shows how PWM works when an analog signal (like audio signal)

“affects” the PWM digital signal. In other words this allows at frequencies of sufficient

frequency to obtain a modulated digital signal of an analog one. Higher the sampling rate or

the frequency of the PWM higher the accuracy of the reproduced digital signal will be. [6]

Figure (1.2) [6] – Analog signal PWM modulated

Analog electronics

Analog circuitry can also be sensitive to noise. Because of its infinite resolution, any

perturbation or noise on an analog signal necessarily changes the current value.


14

Digital control

By controlling analog circuits digitally, system costs and power consumption can be

drastically reduced. What's more, many microcontrollers and DSPs already include on-chip

PWM controllers, making implementation easy.

In a nutshell, PWM is a way of digitally encoding analog signal levels. Through the

use of high-resolution counters, the duty cycle of a square wave is modulated to encode a

specific analog signal level. The PWM signal is still digital because, at any given instant of

time, the full DC supply is either fully on or fully off. The voltage or current source is

supplied to the analog load by means of a repeating series of on and off pulses. The on-time is

the time during which the DC supply is applied to the load, and the off-time is the period

during which the supply is switched off. Given a sufficient bandwidth, any analog value can

be encoded with PWM. [7]

Common modulating frequencies range from 1 kHz to 200 kHz.


15

1.3 Shielding Effectiveness [8]

The need for shielding is because we do not want frequency sources to propagate their

radiation to unwanted places. To prevent occurrence from this, an enclosure must be put around the

radiating device. A schematic picture of this is presented in figure (1.3.1). It is very important to shield

all radiating sources because if not a very dirty environment of radio frequencies will occur.

Figure (1.3) [15]

The shield makes an electromagnetic enclosure of the area of interest. Without the shield there

would exist a vector electric field strength E1 and a vector magnetic field strength B1 at point P. The

vector is fixed in its spatial orientation with its magnitude varying at the frequency f. With the shell at

place the electric field strength at point P will have the new value E2. The electric and magnetic field

at the point P are generally changed in direction and magnitude in the presence of the shell. An electric

shielding effectiveness, SE, and a magnetic shielding effectiveness, SM, are defined as:

= 20 × log

= 20 × log


16

CHAPTER 2 – Platform Design

2.1 Block Diagram

Figure (2.1) – The block schematic of the entire circuit

Voltage Stabilizer 12V

Supply Source 24V ac

~31.5V DC

PWM unit

GDT

Power Circuit

~27 kV

Audio Device Oscillator

Flip/Flop

Output A

Output B

Osc. Probe Osc. Probe

Osc. Probe

Osc. Probe

Osc. Probe


17

2.2 Circuit Schematic

Figure (2.2) – Entire circuit schematic


18

2.3 General Design

The entire circuit design is based on the details provided in each component technical

data sheet. The main integrated circuit (IC) used because of the availability on the market and

also its low cost is a PWM circuit KA3525 the equivalent of the SG3525 also found in

Switched Mode Power Supplies.

This PWM IC has two outputs (pin 11 and pin 14) which can be connected to a GDT

as how figure (2.3.1) states it, to make a MOSFET driver that allows the usage of 2

MOSFETs in a half-bridge configuration in order to minimize the heat sink used. Output

signals are polarity reversed. The GDT is used in order to provide a galvanic insulation

because of the high voltage (HV) electric spikes, which may be reversed by wire in normal

conditions, into the IC.

A voltage limiter of 1A is used to protect the IC from voltage variations.

Capacitors are used in order to filter the DC signal at the input and output of LM also

when leaving unused pins of the IC “in air”.

The MOSFETs commutate DC in the primary coil of a flyback transformer, DC that

being switched charges 2 capacitors that help prevent the arc from extinguishing at lower

frequencies.

In this circuit between the outputs the GDT was connected and a 1 µF capacitor for the

wave correction of the GDT’s primary winding.

When the circuit is started a modified duty-cycle can be obtained by rotating the 10

kΩ potentiometer. By using this input point of the IC, we can connect the analogous signal

that acts itself as a variable resistor due to the wave’s current characteristic to pass below zero

point, making the input’s impedance to rise and fall like the audio signal. This, applied to the

oscillator and PWM part of the circuit prolongs the duty cycle of the PWM, adding to the

already created command at a rate of 37.5 kHz (inaudible arc) the “vibration” of the audio

signal.


19

Figure (2.3) shows the connection of the IC outputs for driving 2 transistors in half

bridge configuration.

Figure (2.3) [9] – Driving transistors into half bridge configuration

Total power consumption at 37.5 kHz:

o ~96 W (3A)

Total power consumption at 15 kHz:

o ~128 W (4A)


20

2.4 The IC

The circuit presented in its block diagram, figure (2.4) is a PWM circuit. The

principles of operation of the PWM circuits and its usages are described in the first part of this

paper.

By applying an oscillation variable in 15 kHz – 37.5 kHz range, to the PWM section

and the Flip/Flop part of the circuit that alternately commands transistors we obtain an

modulation of controlled duty ratio at the two outputs that are phase-shifted.

Figure (2.4) [9] – IC block diagram


21

Aside from chosen capacitors and other external elements in this circuit, the PWM

circuit has an internal oscillator that is calibrated using the formula:

f = 1

CT (0.7 RT + 3 RD)

Resulting operating frequency for this circuit is f = 37.5 kHz.

The approximate charging times and values for the driving components can be

obtained from figure (2.4)

Figure (2.4) [9] – Calibration of oscillator frequency using RT and CT


22

2.5 GDT

These devices provide electrical pulses for turning on and off semiconductors, such as

high-voltage power MOSFETS or IGBTs. They also are used for galvanic isolation. Gate-

drive transformers are essentially pulse transformers that are used to drive the gate of an

electronic switching device.

For the presented circuit a ring shaped ferrite core GDT 1:1:1 ratio was used due to

high frequencies used to obtain the arc.

The number of turns for the GDT was determined experimentally following a table

containing a typical set of incorrect waves that can be obtain when designing a GDT.

Usually when all details provided and detailed datasheets are available, a

mathematical calculus is more efficient when designing a GDT but when working with a

variable length electric arc, also generating a wide range of frequencies like in present case,

the most efficient and fastest way to determine the on-circuit parameters is to make use of the

workshop equipment and modify the GDT parameters considering the gathered information.


23

2.6 GDT Waveform Correction Table [7]

The perfect waveform

- Flat tops and bottoms to the pulses

- Steep rising and falling edges

- Little overshoot at the switching

transitions

- No ringing after the transitions

Good gate drive waveform except for

overshoots

- Flat tops and bottoms to the pulses

- Steep rising and falling edges

- Considerable overshoot at the

transitions due to insufficient damping

resistance

- Some ringing after the transitions, but

not too bad.

Solution:

Increase series damping resistor slightly and

the overshoot should diminish.


24

Poor gate drive waveform with

excessive high-frequency ringing

- Large overshoot at the switching

transitions

- Prolonged high-frequency ringing due

to lack of any damping resistance

- Totally unusable because ringing takes

the MOSFET repeatedly into its linear

region

Solution:

Adding some damping resistance.

Reducing leakage inductance if possible.

Slightly over damped gate drive waveform

- Leading edges of pulses are curved due

to too much damping resistance

- Slow rise and falling edges mean

MOSFET spends longer than necessary

changing state

- This causes heating due to high

switching losses

Solution:

Decreasing the damping resistor, to make the

rising and falling edges faster


25

Massively over damped gate drive waveform

- Waveform looks like "shark's fins"

because of far to much damping

resistance

- Can also occur if the driver is totally

inadequate to drive high gate

capacitance

- Totally unusable because MOSFETs

would spend all their time in the linear

region

- This causes rapid overheating of the

MOSFETs

Solution:

Decreasing the damping resistor, or usage of a

more powerful gate drive IC.

Slightly sloping tops and bottoms to waveform

- The tops and bottoms of pulses droop

slightly towards zero

- Caused by low primary inductance.

Too few turns on the drive transformer

- This is not a problem as long as the

amount of droop is less than a couple

of volts

Solution:

Add a few more turns to the primary and

secondary windings to reduce the droop.


26

Excessively drooping tops and bottoms to

waveform

- The tops and bottoms of pulses slope

steeply down towards zero

- Far too low primary inductance. Either

too few turns or wrong core material

used

- Unusable because the MOSFETs

would start to turn off towards the end

of the pulses

Solution:

Use many more turns or choose a core type

with higher Specific Inductance.

Excessive low-frequency ringing

- Severe low frequency ringing due to

excessive leakage inductance in drive

transformer

- Totally unusable because ringing takes

the MOSFET back into its linear

region

Solution:

This cannot be corrected by increasing the

damping resistance. It would need too much

resistance, and the rising and falling edges

would be too slow. Must reduce the excessive

leakage inductance by redesigning the GDT.


27

Resonant gate drive, (Sinusoidal)

At very high switching frequencies it is

possible to make use of the resonance caused

by the leakage inductance and MOSFET gate

capacitance. This technique is used to good

effect in RF amplifiers that operate in the

switching mode up to tens of Megahertz. At

first a sinusoidal waveform may not seem

ideal for driving a MOSFET gate. However, it

does have moderately fast rising and falling

edges where the sine wave passes through

zero.

This technique is only mentioned here for

completeness. It is not commonly used below

a couple of Megahertz. Square wave drive

always yields lower switching losses when

possible.


28

2.7 GDT calculus

Q factor was not of importance herein therefore a common ferrite core of ~ 640 µr was

used.

Ferrite dimensions:

- outer diameter 25

- inner diameter 15

- length 12

A 0.25 mm copper wire radius was used.

Number of turns – 25

Circular loop inductance formula:

L≅ 8

− 2

Where:

- N – number of turns

- R – radius of circle [m]

- A – wire radius [m]

- – relative permeability of the medium

- L – inductance [H]

For an inductor the formula for impedance calculus is:

Z = × 2 × × ×

- j = the square root of -1, and accounts for the phase relationship between the

impressed voltage and resultant current.

- f = frequency of applied voltage [Hz]

- L = inductance in Henries


29

An inductor using a core to increase inductance will have losses such as hysteresis and

eddy current losses in the core. At high frequencies there are also additional losses in the

windings due to proximity and skin effect. These are in addition to wire resistance, and lead to

a higher equivalent series resistance.

By applying the 2 above listed formulas for an ideal case we obtain:

- L = 0.00002507 [H]

- Z at 37500 Hz is somewhere around 5.89 Ω

2.8 Short guide and references for designing a GDT [8]

1 core material selection based on operating frequency that will determine the amount of

inductance that is needed on the primary of the gate driver transformer

2 minimizing parasitic influence by using one of the available formulas for estimating

leakage inductance in the magnetic design (avoid having half turn winding because this

leads to leakage inductance)

3 calculate the number of turns used

4 calculate the wire gauge


30

2.9 The Flyback Transformer

A flyback transformer is a DC-DC converter. It takes a small voltage at a small

number of windings on the primary coil and creates a large voltage on a large number of

windings on the secondary coil. This works because when the low voltage is applied at the

primary coil, it is at a fairly high current ~3A, and creates a magnetic field. When the current

is removed, the magnetic field collapses in on the transformer core and all the energy is

dumped back into the windings of the transformer. This electrical energy is transferred to each

winding, and since the secondary coil has so many windings, the voltage created is very high.

When the magnetic field is collapsing and the energy is being dumped into the coil, the

transformer is said to be in a state of flyback, hence the name. [14]

In this circuit a frequency above 20 kHz was necessary to obtain an above hearing

frequency arc and therefore a flyback transformer operating typically at around 15 kHz but

capable depending on the ferrite core to withstand switched inputs up to 50 kHz was used.

Experimentally was determined that the saturation point for the transformer used in

this circuit is at around 37 kHz.


31

2.10 Unprotected circuit waveforms

When working on the test board and not using grounding many of the wires act as

antennae. This affected the circuit by introducing noise into the supply wires therefore

affecting the output of the circuit and MOSFETs command.

Figure (2.5) shows the sinusoid waveform measured at the secondary of the

transformer with the main HV circuit unplugged. The power consumption was determined to

range between 200 mA and 400 mA depending on the frequency of the oscillator and was

observed to increase with the number of turns on the GDT which is connected between the

outputs of the IC.

Figure (2.5) – Sinusoidal waveform of the output of 24V transformer

We can observe that, because of the high current charging drawn by the two 10 mF

capacitors, the voltage slowly drops due to incorrect (low value) capacitance used for the

capacitor placed in parallel with the rectifying bridge for the circuit. The capacitor charging

current is depicted in figure (2.6) and for this circuit has an initial value of around 6A for a

period of 0.04 s. Figure (2.7) shows the rectified voltage, with the HV circuit running, of the


32

supply source that has a value of approximately 31.5 V; exact mathematical value can be

determined by measuring the rms value at the ac output of the transformer multiplied by √2.

Figure (2.6) [10] – Diagram for calculating the instant charging time of capacitors

Figure (2.7) – Rectified voltage with running arc


33

Figure (2.8) – Shows the no load DC rectified voltage

In figure (2.9) we can see that because of the high charging current for the two 10 mF

capacitors also the sinusoid waveform is affected (notice the distortions at the peak values of

the sinusoid). This also causes some perturbations in the network due to rapidly withdrawn

currents from the transformer.

Figure (2.9) – 2 X 10 nF Capacitors drawing high current when charging


34

Figure (2.10) shows the first probe of the oscilloscope placed on the voltage limiter

output (output not filtered with the 1000 µF capacitor). Due to the switching state of the

circuit and the radiated electromagnetic fields of the arc we can notice severely captured

peaks in the supply DC. These spikes appear when the transistors conduct and turn off and

have a frequency equal to the oscillator frequency. This needs to be avoided because due to

the varying of the DC the output amplitude of the IC changes and the sound of the modulated

signal picks up the noise created by the switching state of the transistors.

Figure (2.10) – Input voltage without decoupling in contrast with transistor’s command signal

Figure (2.11) shows the previous explained output of the voltage limiter filtered by the

1000 µF capacitor connected on the schematic between the output pin of the voltage limiter

and the ground respecting polarity. However we can see the second probe connected to the

input of the limiter and also the admission of the spikes in the rest of the bridge rectifier

circuit. This is to be corrected due to the lower amplitude of the spike than the previously

studied spikes with a 0.1 µF capacitor. This also affects the normal operation of the circuit

and IC.


35

Figure (2.11) – Input voltage with decoupling in contrast with input voltage without

decoupling

Figure (2.12) shows the first probe connected to the output of the IC and the signal

almost a perfect square. These tiny variations are present due to the poor quality of the

oscilloscope graphics. The second probe is attached to the output coil of the GDT where we

can observe that some spikes appear at the transitions between the ground level and the peak

voltage of the square signal. These perturbations are present in here because of the low value

series resistor at the output coil of the GDT. This can easily be corrected by adding some

surplus resistance in order to diminish the overshoots. However MOSFETs work with this

gate command too but if we need precisely modulation of the sound and highest quality we

should design the GDT as to be perfect flat tops and bottoms. Also we reduce also the heating

of the MOSFETs by giving them a correct command waveform.


36

Figure (2.12) – Output signal from IC in contrast with signal from GDT output

Figure (2.13) shows a close-up image of the perturbations created due to the incorrect

GDT that in this case can be corrected by adding some damping series resistor that increases

the impedance of the circuit.

Figure (2.13) – IC output signal in contrast with GDT output signal


37

Figure (2.14) shows a slight perturbation in the command signal of the MOSFETs at

the switching frequency of 37.5 kHz in the channel 1 of the oscilloscope. The “line”

represents the continuous current that also suffers very little from the effect of the

perturbations appeared in the first ns after the switching point perturbations not very visible

due to protections inserted into the circuit discussed and presented in the figure.

Figure (2.14) – Perturbations in the command signal and also into the protected input voltage

at highest frequency

Figure (2.15) shows the perturbation that we can visualize on the oscilloscope that

appears at the lowest frequency this circuit can reproduce. It is an obvious rise in the time

period of the perturbation once the frequency got lower.


38

Figure (2.15) – Perturbations in the command signal and also into the protected input voltage

at highest frequency

Figure (2.16) shows the oscilloscope probe 1 connected after the GDT and the second

oscilloscope probe connected before the GDT. We can observe that due to the incorrect

wiring of the GDT the output signal (1) is 2-3 times bigger in amplitude than the input signal.

This happened because the GDT is manufactured by hand and the number of turns was not

preserved exactly. At these high frequencies even a half of turn is responsible for great

deviation of the signal and depicted below is the measured proof of this theory.


39

Figure (2.16) – IC signal before GDT and after

In figure (2.17) channel 1 shows the signal at one of the GDT outputs with the series

22 Ω resistor depicting the signal as clear as it gets with the running arc in contrast with the

2nd probe of the oscilloscope connected to the same output of the GDT but without the series

damping resistor connected.

Figure (2.17) – Signal with and without 22 Ω damping resistor


40

Figure (2.18) shows the correctness of the two IC outputs that are polarity reversed.

Figure (2.18) – Corrected GDT signals


41

2.11 Circuit parts

- 230 V ~ 630 mA to 12 V ~ 6 A and 24 V ~ 3.3 ferromagnetic core transformer

- 2 rectifiers

- 1 ka3525an (16 pin PWM IC)

- 1 2.2 µf capacitor

- 1 2.2 k resistor

- 10 k potentiometer

- 1 3.3 µF

- 2 x 0.1uF non-polarized capacitors

- 1 LM7812

- 1 1000 µF capacitor

- 1 1 µF

- 2 x 22 resistor

- 2 x IRFP250 transistor

- 2 x 10 mF polarized capacitors

- 1 flyback transformer

- 1 GDT ~ 27 turns, 3 wires



- 3 heat sinks

- 2 inox steel electrodes

- wires

- 1 PCB with the command part of the circuit depicted above

- soldering iron

- wood

All of the above listed components were put together by taking into consideration existent

risks of component explosion in case of incorrect wiring of the circuit.


42

2.13 PCB Construction

Figure (2.19) – Ready to print PCB

The PCB board comes coated in Cu on either one side or both. After designing the

circuit in some software program (plenty available) the schematic gets laser printed on heat

resistant paper after which by thermal transfer the blueprint of the circuit gets imprinted on

the copper plate. Next part of the process is to dip the plate into ferric chloride 40% solution

for a period of approximately 20 minutes, time in which the excess and unprinted copper

reacts and dissolves.

2.14 Faraday’s Cage

A Faraday cage or Faraday shield is an enclosure formed by conducting material or by a mesh

of such material. Such an enclosure blocks out external static electric fields. Faraday cages are

named after the English scientist Michael Faraday, who invented them in 1836. [5]

This is a necessity in this circuit due to high electric, magnetic and electrostatic field created

by the high voltages when arming the arc.


43

2.15 MOSFETs

Although the MOSFETS work in a half bridge configuration and alternately turn on

and off the do need heat sinks because of the high current switching. The heat results from the

power losses.

We take a quick look at MOSFETs, particularly and how they work. In basic terms,

the MOSFET is a voltage controlled current source, with a voltage difference from gate to

source causing a current to flow from drain to source.

N-channel MOSFET schematic symbol:

Figure (2.20) – MOSFET circuit representation

Varying the gate voltage will vary the current in the device’s drain.

The following table (2.1) contains values from the MOSFET’s datasheet:

Gate-Source Voltage Operating Mode

+VGS – max and above Gate breaks down, device damaged

+4V to VGS – max On

+4V to +2V Linear

+2V to –VGS – max Off

-VGS-max and below Gate breaks down, device damaged

Table (2.1) [8] – Operation modes of the MOSFET


44

When ON, the MOSFET appears as a low value resistor (typically less than 1 ohm)

shown in the datasheet as the RDS-ON value (on resistance from drain to source), which is

also dependant on temperature as well as gate voltage. In this low resistance state a large

amount of current can flow through the device.

When OFF, the device appears as a high value resistor and very little current flows

from D -> S. We ideally want to keep the device operating in either the ON or the off states.

When in the linear region, the MOSFET acts like a resistor and can dissipate large

amounts of power. For switching circuits, we want to avoid operating in this region as it

causes heating of the device.

Gate voltages are given limits in the device datasheets (VGS – max), usually +/– 15V

relative to the source. Exceeding this voltage can damage the device, causing a short circuit

between gate and drain (or source).

Often in designs, back-to-back Zener diodes can be mounted across gate-source

terminals to protect against over voltage on the gate. If you have a good GDT design, these

should not be necessary, although some people include them to be safe. [13]


45

2.16 Physical construction

The circuit began as a sum of cables connected on a test platform, cables which acted

as antennas and gave interference into the circuit as well as connecting measuring apparatus

to the key points. In figure (2.21) we can see the laying cables on a test table.

Figure (2.21) – First state of the circuit


46

Figure (2.22) shows the PCB after the corrosion process in FeCl3. The process consists

in designing the circuit using available software, printing on a single or double sided copper

coated plate and then sunk for a period of approximately half an hour in FeCl3.

Figure (2.22) – PCB after corrosion


47

Figure (2.23) shows the PCB after mounting the command circuit parts. These parts

were separated on a PCB from the rest of the circuit because none of these components

needed active cooling so the length of the paths was maintained short in order to diminish

their effect as antennas. In this picture we can observe the pins which were separated from the

circuit making room for the oscilloscope probes to be attached.

Figure (2.23) – Fixed components on completed PCB


48

Figure (2.24) shows the running arc enclosed in a metal grounded Faraday Cage used

for protection of the external electromagnetic fields and also not to radiate in the environment

electromagnetic waves.

Figure (2.24) – Faraday caged electric arc


49

CHAPTER 3 – Optimization of the Circuit by Different Means of Protection

1.1 Cable Shielding [10]

A shielded or screened cable is an electrical cable of one or more insulated conductors

enclosed by a common conductive layer. The shield may be composed of braided strands of

copper (or other metal), a non-braided spiral winding of copper tape, or a layer of conducting

polymer. Usually, this shield is covered with a jacket. The shield acts as a Faraday cage to

reduce electrical noise from affecting the signals, and to reduce electromagnetic radiation that

may interfere with other devices. The shield minimizes capacitively coupled noise from other

electrical sources. The shield must be applied across cable splices.

Microphone or "signal" cable used in setting up PA and recording studios is usually shielded

twisted pair cable, terminated in XLR connectors. The twisted pair carries the signal in a

balanced audio configuration.

1.2 Transformer Shielding

The supply source which is a transformer should also be shielded. Being in the vicinity

of the low powered unprotected circuit this component acts as a perturbation source inducing

the 50 Hz sinusoidal wave especially into the GDT which is also a 1:1:1 transformer.

1.3 Shorten Paths

By printing the circuit board, the length of the paths was minimized allowing an

ammeter to be connected without creating audible perturbations in the arc.


50

1.4 Hazards

Important information regarding hazardous connections made in the circuit

- Never make connections between the transistor’s heat sinks because the common

connection will be common drain; this will create a short circuit at around 3A and will

probably affect the supply source due to the bypass fuse which has a greater value than

recommended 0.63A therefore allowing the flow of greater currents in the

transformer’s secondary winding

- When using a portable apparatus which has a radio AM/FM function do not create a

connection between the apparatus and the cage or the heat sinks and the apparatus

because spikes of high voltage run through this components; part of them will be

radiated in air and part of them will flow through wires and the high sensibility of the

apparatus’ antennae will allow a short circuit between these parts to be created

therefore destroying the apparatus

- When working al lower frequencies a time of approximately 30 s should not be

overdue because the end of the electrodes will get very hot and eventually deform due

to the excess heat

- Under no circumstances should be made connections on the circuit exposed other than

connecting the probes of the oscilloscope to the specially designed pins.

- If cooler is malfunctioning the device should not be turned on for longer than 1 minute

- Powerful short-circuit will be made if for example the audio jack is touched by either

one of the radiators or other metallic parts


51

- The grounding circuit with respect to other components contains strong currents and

therefore between this part of the circuit and other components or other apparatus

should not be made any connections for operator’s safety and apparatus’ as well

- The electric HV arc closes between pin 5 and the HV wire coming at the other end of

the flyback transformer therefore other connections of one of its pins is useless and

therefore the risk of electrocution is only present at the connection of the two

electrodes

- Powerful heat is transferred through convection from the arc especially at low

frequencies to the surrounding air. Temperatures can reach 1000+ Celsius degrees


52

CHAPTER 4 – Laboratory Project

4.1 Laboratory Project Proposal

By making use of the built platform in a controlled environment I can propose some

laboratory project titles like:

- Study of the coupling and decoupling of the arc over the oscilloscopes

o This can be achieved by making use of the frequency potentiometer to give the

arc a lower frequency so that it can be blown up by a student.

It will be studied the interference obtained when arming and disarming

the arc seen even on an oscilloscope’s CRT.

- Study of the interference produced by the switching transistors

o This can be achieved by placing oscilloscope probes at the gates of the

MOSFETs and observing the perturbations which appear at the rising and

falling edges of the modulated pulses.

- Study of the electrostatic field (without Faraday’s Cage) created by the 15 kHz blown

electric arc

- Study of the conducted interferences and difference between them by connecting and

disconnecting the protective Faraday’s Cage from the grounding system.


53

Conclusion

After this laborious work we can conclude that all equipments especially power

equipments and switching devices generate perturbations which are more likely to be

conducted perturbations than radiated ones.

Most affected are the equipments that are not provided with a grounding system or a

protective cage in case the elements contained in the circuits have coils capacitors or long

connection paths.

After the study of this circuit we have seen that protection exist ranging from the

simplest electronic components like capacitors or capacitors connected with resistors creating

filtering either high pass, low pass or band pass. In addition to this presented elements, some

other exist:

- Ferrite Beads

- Faraday’s Cage

- Shielded cables

By having a relatively low impedance of the GDT we can assure a quality

transmission of the output signals to the gates of the MOSFETs therefore optimizing their

operation by steeply rising the command signal to the open gate to source ON voltage

eliminating losses by means of heat dissipation.

As an audio speaker we can notice that the quality of the signal is very high,

practically not existing any mass in the speaker the frequency response is instant.

All of these performances are obtained at a high power cost therefore we can notice

the very low sub 0.1 efficiency of the circuit.


54

Bibliography

1. http://searchmobilecomputing.techtarget.com/definition/electromagnetic-interference

2. http://electronicsbus.com/emi-electromagnetic-interference-rf-noise-radio-frequency-

interference/

3. http://en.wikipedia.org/wiki/Electromagnetic_interference "summary about EMI

word"

4. Conf. Dr. Ing. ACIU LIA, "introduction in EMC"

5. "Faraday, Michael - ninemsn Encarta". Archived from the original on 31 October

2009.

6. http://www.acroname.com/robotics/info/concepts/pwm.html

7. http://www.netrino.com/Embedded-Systems/How-To/PWM-Pulse-Width-Modulation

8. IRFP250 datasheet

9. SG3525 datasheet

10. http://hyperphysics.phy-astr.gsu.edu/hbase/electric/capchg.html

11. http://www.richieburnett.co.uk/temp/gdt/gdt2.html

12. A Guide to Designing Gate-Drive Transformers; By Patrick Scoggins, Senior

13. http://thedatastream.4hv.org/gdt_gate.htm

14. 14 Dixon, Lloyd H, Magnetics Design Handbook, Section 5, Inductor and Flyback

Transformer Design, Texas Instruments, 2001

15. Shielding Methods for Radio Frequencies, Anton Brink; Department of Electroscience

Electromagnetic Theory Lund Institute of Technology Sweden, 2001

16. http://www.boeing.com/commercial/aeromagazine/aero_10/loop_textonly.html

17. http://hyperphysics.phy-astr.gsu.edu/hbase/electric/capchg.html


Appendix

List of used abbreviations:

GDT – Gate Drive Transformer

HV – High Voltage

IC – Integrated Circuit

HF – High Frequency

PWM – Pulse Width Modulation

LM – Limiter

MOSFET – Metal Oxide Semiconductor Field Effect Transistor

PCB – Printed Circuit Board

CRT – Cathodic Ray Tube

Date post:	02-Oct-2014
Category:	Documents
Upload:	bogdan-gotia
View:	69 times
Download:	0 times

finalul

Documents