Design and Simulation of Class D Audio Amplifier
Dissertation
Title: Class D Audio Amplifier
Abstract:
The dissertation about designing and simulating of a class D digital audio amplifier, which is usually use in different applications in human regular life such as audio players, tablets, headphones, smart phones. Class D amplifiers available from 1960’s on wards in the field but not regularly used in large fidelity applications. Class D amplifiers also known as switching amplifiers. A class D audio amplifier normally have high efficiency for any range of output power from medium to high. The class D amplifier naturally have compatibility respectively with pulse width modulation and nearly 100% efficiency. Class AB amplifiers give very high excellence audio but fail in efficiency comparison with class D amplifiers, those amplifiers generally low efficiency. Class D amplifiers uninterruptedly switch output from individual railing to the further at supersonic frequency, controlling the space ratio to give an average representing the instantaneous level of the audio signal. This is alternatively called Pulse Width Modulation. The input audio frequencies here using range about 10 Hz to 20 kHz. The received audio signal desires to be trained and filtered earlier it compared with triangle wave. And low-pass filter must be used for stop aliasing, then the level must be limited less than that of triangle wave. And the amplitude in audio signal could need to stand attenuated or else amplified for contest that with comparator supplies also with triangle wave amplitude. For increase signal-to-noise ratio, then peak level audio of input must be as near to system and full scale as much as possible. Dependents on the particular application and also loudspeaker will be driven, and it could be useful for band limit of input signal. good example, when a small speaker used that really cannot deliver tones that less than 100 Hz, and the input must be high pass filtered for reduce lost energy and there have possible for speaker damage.
Contents
What are audio amplifiers for?
Efficiency of Class-G amplifier:
Basic MOSFET Structure and Symbol
Why Filter is required in Class-D Amplifiers:
Topology Comparison – Linear vs. Class D
The Class D Amplifier Advantage
Table of figures:
Figure 1: Class-A Audio Amplifier
Figure 2: Class-B audio Amplifier
Figure 3:Class-AB Audio Amplifier
Figure 4:Basic Block diagram of Class-D audio Amplifier
Figure 5: Output Waveform from Class-D audio Anplifier
Figure 7:Basic Operational Amplifier
Figure 8: Basic Class-D amplifier Low pass filter
Figure 9: Topology Comparison – Linear vs. Class D
Figure 10: Circuit diagram for PWM wave
Figure 11: Triangular and sinewave frequencies
Figure 12: Pulse Width Modulation signal from Operational Amplifier
Figure 13: Switching stage of Class-D audio Amplifier
Figure 14: Output signal from the Switching Stage
Figure 15: Filter design for amplifier
Figure 16: Audio output signal
Introduction:
The primary theme of audio amplifiers stands to give return of audio signal input into sound producing output at proper power levels and desired volume levels having efficiently at low distortion. The amplifiers must produce a very good frequency output between the ranges of 20 Hz to 20 kHz. Depending on the application the power levels may vary in headphones it may be in milliwatts, in TV or PC it may be in watts, in home stereo and automotive it may be vary in tens watts and in commercial sound systems it may be vary more than hundred watts. So in order to fill sound in theatres and auditoriums the audio amplifier with analogue implementation using straight linear transistors were used to create output voltage respective to input voltage. Here we find a very high level of forward gain voltage, if it is a part of loop obviously the gain will rise up to high in overall loop. Why because always high loop gain always improves the performance at distortions which were caused by nonlinearities in forward paths and reduces the noise of supply by power supply rejection which increases. (Slone, 1999)
What are audio amplifiers for?
In sound structure, power amplifiers stand the bridge amongst the loudspeaker plus the remaining of other sound system. In ordinary idiom, ‘Audio Amplifier’ becomes abbreviated of ‘amplifier’ and also ‘amp’. But altogether audio amplifiers (further than individuals that determination disc cuter heads) stand truly ‘loudspeaker drivers’. And the meaning is global uncertainty earpieces in addition headphones are involved. Sometimes, amps remain combined by the speakers, creating ‘powered speakers’; otherwise they might be packed with the forgoing equipment purposes, e.g. by way of in national ‘integrated’ hi-fi kind amplifier or +preamplifier, and also band’s ‘mixer-amplifier’. (Duncan, 1996)
Class- A Audio Amplifier:
Class-A Amplifier is one of the simple device among compared with other amplifiers. This amplifier well named for low level signal distortions and gives the best output in sound quality, the simple and great reason to select class amplifier is for highest level of linearity operating at linear portions of the characteristics curve. (Watkinson, 2001)
Since the amplifier class A uses single transistor like Bipolar, FET, IGBT etc. which is connected commonly to emitter configuration to both halves of waveform with always current flowing through transistor since without having base signal also which is nothing but the output phase of Bipolar, MOSFET, IGBT can never go to the state of cut-off or even saturation region also, so unsteadily it will pass through base biasing point Q in the middle of load line. So the transistor could never be turned off or fails which is a great act. (Self, 2006)
Figure 1: Class- A Audio Amplifier
To obtain high linearity and gain the output of class A is biased ON in all time, since it is classified as class A due to the output stage of zero signal idle current must be equal or must be greater than the load current which can be used to produce biggest output signal
Class A is equivalent to current source because it will operate the linear portion of the characteristic curve and signal output conducts a full waveform of 360 degrees output.
Here our class A is being operated in linear region, the voltage of transistor base biasing have to be selected for exact operation and low distortion and at all times the output is ON which carries the current continues sly which states the amplifier is continues loss of power
It creates so much heat because of power supply flows continue sly, adding low efficiency at the rate of 30% making them not to work out for high power amplifications and also due to high current idling the power supply is sized as per the filter to avoid the number and other distortions. So this Class- A amplifier is more efficient due to low efficiency and overheating problems. (Self, Handbook, Audio Power Amplifier Design, 2009)
Class-B audio amplifier:
When input signal conducts positive, the transistor conducts into positive biased and negative transistor goes into OFF state, likely when input signal conducts negative the positive transistor goes into OFF state and turns ON the Negative biased one so negative signal has been conducted. So the transistor conducts only half at time weather it is on positive or negative of the input signal
Here we can observe that the class B transistor conducts only half i.e. 180 degrees giving output as waveform. Here the output stage has signal on both halves of the side and these two halves combines and emits a signal of full linear output waveform
Figure 2: Class-B audio Amplifier
The design of push pull amplifier is much efficient than class A up to some extent here the main problem with class B amplifier is at zero crossing point waveform this can create distortions this is due to the input voltages transistor from negative 0.7 to positive 0.7
As we discussed from tutorials that it takes 0.7 volts from base emitter voltage in order to obtain a bipolar transistor and to conduct that. Here the output transistor is not biased to ON state in
Class-B amplifier until the voltage limit is crossed.
That’s nothing but the waveform falls within 0.7 volt can’t be generated in order to make class B audio amplifier for the further applications
So in order to remove these kind of distortions Class AB were introduced
Class-AB Amplifier:
The name itself states that this amplifier is the made up of combining both class A and class B amplifier. Now a days it is most currently and commonly used amplifier for the purpose of audio power amplification. Here as we said that this Class AB has a variation compared with the above class B amplifier except that variation these both amplifiers conducts at same time at the waveforms crossover points removing the distortions of the crossover problem of class B amplifier
The bias voltage of these two transistors is very small which is between 5 to 10 percent to bias the transistor at quiescent current which is slightly above the point at cut-off. So this conducting device becomes ON state at either FET or Bipolar which is at more than the level of half cycle of the input signal. Here in this amplifier the transistors both push and pull conducts more than half than the conduction in class B and less than the full cycle conduction of class A.
Which means here the conduction of class AB lies in between 180 degrees and 360 degrees, which depends upon the situation as we can see in figure.
Figure 3: Class- AB Audio Amplifier
The main advantage is only that to overcome the crossover distortion which is created by class B, which is obtained by a small biased voltage connected in series by diodes and resistors without taking care of negative effects of Class- A amplifier. So in treating for efficiency class AB amplifier is very much better compared of class A and class B amplifier reaching the conversion energy about 50 percentage to 60 percentage.
Class-G amplifier:
Most category audio amplifier a smaller amount efficient compare to Class-B; for example, first Class-AB is distinctly less efficient on the low close of the aforementioned power capability, though it is pure that Class-A trashes virtually completely the energy place into it. Constructing amplifiers with advanced efficiency is extra difficult. Class-D, spending ultrasonic pulse-width modulation, capacities high efficiency also sometimes smooth delivers it, then again it is incontestably a problematic technology. The real efficiency of Class-D breaks on facts of circuit strategy and device individualities. The apparently unescapable LC output filter-second order on smallest – can only provide a smooth response into unique load impedance, then its magnetics remain neither reduced nor easy to project. There are possible to be around daunting EMC problems through emissions. Class-D stands not a smart proposition used for superior domestic amplifiers that necessity with discrete speaker of unidentified impedance characteristics.
Efficiency of Class-G amplifier:
The normal mathematical descent of Class-B efficiency through sinewave drive expenditures straightforward integration concluded a half-cycle toward calculate interior dissipation compared to voltage fraction, i.e., the portion of possible production voltage swing. As is sound known, now Class-B the supreme heat dissipation stands around 40% of determined output power, on output voltage portion of 63%, that means which too transports 40% of the concentrated output power towards the load. The arithmetic is simple since the waveform seriously do not differ in shape through output level. Each possible idealization remains assumed, such by way of zero inactive current, not any emitter resistors, also no more Vce losses then so on. In the Class-G, arranged the extra hand, waveforms remain a strong task of at output level, demanding variable parameters of integration also so on, in addition to it all becomes very heavy. Yielding power partition drawing, which displays in what way the power strained from the source is distributed among output device degeneracy and valuable power on the load. No one disagreement that sinewaves stand poor simulations for music used for this purpose, then their main benefit, that they tolerate direct contrast with the morally mathematical method. However, meanwhile the total point of a Class-G is to power saving, also the waveform recycled has a solid result on the outcomes.
Class-D amplifier:
Class D amplifier consume increased extremely in acceptance. This is due to class D amplifier gives the maximum efficiency in any kind amplifier classes, while the presentation, particularly popular in standings of linearity, and is not so decent.
The field of solicitation for this class D amplifier be able to be approximately divided into dual parts; high and low power outputs. Here low power area reaches starting from a partial milliwatts onwards (for digital range aids) toward around 5W, though the great power applications will go starting 80W – 1400 W. At current there performs to be approximately of a break in the mid, for purpose that will appear. The low-power field contains applications such for example personal stereos, mobile phones, computer audio and laptop. These yields are movable, and battery energetic, so the power economy very important. And the major application in class-D is the manufacture of quantity of audio power as of a particular low-voltage source rail. A very good example stands the national semiconductor LM4671, a solo-channel amplifier integrated circuit (IC) that give around 2.1W through a 4 ohm speaker beginning around 5 V supply rail, consuming 300 kHz of switching frequency. And this is a too low voltage through straight power amplifier principles, and necessitates an H-bridge structure of output, of which extra later.
The application of power amplifiers includes PA amplifiers, home theater systems and big sub-woofers. From the main supply these are all reinforced, hence power doesn’t keep any tremendous priority. Since class-D is self-indulgence and there will be a loss of minimum power supply and heat sink size which results mini and perfect output product. In car audio systems we are going to use high power audio systems, having power capabilities more than 1000W or 2 ohm .Here we give importance in minimizing the power drain, providing 12v capability when engine is driver to alternator are finite. Whereas in these two areas a middle ground exists, when an amplifier have fed from the main supply we cannot find such a good and proper output – let us assume a stereo dissipating 30 W output having 8 ohm per channel . Here you can observe small heat sinks, in order to exist the product in our budget removing ios not our main theme.
History of Class-D amplifier:
Coming to the point of technology we may think that the history of the class D amplifier. In 1950 the principle is generally discussed since the combination of the valued output transformers and high switching value frequencies were not much tempting. Sinclair X10 in UK is the first born of class D amplifier resulting an output of 10w, due to the rudimentary output filter followed by THD 5% which did very well keeping low switching frequency followed by load. At that time the common and bad problem are the power handling capacity of the bipolar transistors were poor to switch the frequency when required which is a major drawback in class D producing too much distortions with switching times not the power FETs from then onwards class D has become much propositional.
Basic principles:
Compared to the classes of A B and G class D is little bit much different. In linear mode no output devices are there in class D. They has been operated to switch ON and OFF at ultrasonic frequency. To the each supply rail the output has been connected, when the limit of the input signal varies we can observe the variation of output voltage also due to low pass output filter or by loudspeaker inductance we will perform the averaging, here we have to remind that output voltage is directly proportional to supply voltage having nil inherit rejection of supply at output stage where as in class B output stage we can find negative feedback of range between 50khz to 1mhz of switching frequencies. A high frequency makes output minimal and less complex but gives rise to loss of switching and distortion. By using the differential amplifier we can generate the drive signal which is olden method. One input is received by incoming audio signal and other by the triangle waveform at particular time and frequencies. (Hood, 1997)
Figure 4:Basic Block diagram of Class-D audio Amplifier
A basic Class-D amplifier is as shown in the above figure and the PWM process is illustrated in below figure
Figure 5: Output Waveform from Class-D audio Anplifier
Clearly in order to eliminate the distortion a linear triangular wave is used. We can use other ways also like sigma delta modulator. Our point is to generate much and much amount of audio power with low supply voltage like 5V integrating a H-Bridge configuration which allows to swing the voltage twice across the load and four times of power theoretically admits the amplifier to tune out without output capacitors from the supply rail. This process is known as Big Tied Load (BTL). (Singmin, 2000).
MOSFET:
The MOSFET means Metal Oxide Semiconductor Field Effect Transistor, has a very great input at gate resistance through the current rolling through the channel among the source then drain being well-ordered by from gate voltage. Due of this very great input impedance and also gain, MOSFETs can stand simply damaged from static electricity in case not sensibly protected. The Metal oxide semiconductor field effect transistor are perfect for custom work electronic switches and also common-source amplifiers for power consumption is too small. Very Typical applications of metal oxide semiconductor field effect transistors in Memories, Microprocessors, Logic CMOS Gates and Calculators etc.
And also, in below basic diagram poster that a broken line inside the symbol specifies a generally “OFF” enhancement kind presentation there is “NO” current can possible to flow over the channel after zero of gate-source voltage (VGS) is applied. Here continuous steady line inside the mark indicates a very normally “ON” case of Depletion type display that particular current “CAN” flow over the channel also with zero level gate voltage. For p-channel kind the symbols stand exactly the similar for both forms excepting the arrow themes outwards. This also can be summarised at the below switching table.
| MOSFET type | VGS = +ve | VGS = 0 | VGS = -ve |
| N-Channel Depletion | ON | ON | OFF |
| N-Channel Enhancement | ON | OFF | OFF |
| P-Channel Depletion | OFF | ON | ON |
| P-Channel Enhancement | OFF | OFF | ON |
Here n-type of enhancement type MOSFETs, and positive gate of voltage “ON” the transistor then at zero gate voltage transistor will turn “OFF”. And P-channel enhancement of MOSFET, a -Ve gate voltage turn “ON” in transistor and also by zero gate voltage, here transistor condition will “OFF”. Voltage point of which this MOSFET starting point for pass current over the channel remains determined through the threshold voltage VTH for the device.
Field Effect Transistors
Till now we used transistors as amplifying devices, at solid state switch in saturation and cut off regions we took the operation of transistor and in order to switch the DC current On, OFF we had taken field effect transistors like LED’s which consumes a little bit of milliamp current
Basic MOSFET Structure and Symbol
Here compared of junction FET metal oxide semiconductor FET is quite different. The electrical fields produced at gate level voltages were used by MOSFET’s, the electrons of n- channel and holes of p- channel always altered by the flow of charge carriers by gate voltage through drain, in a very thin insulating layer on top a gate electrode is placed and under the drain and source electrodes there were a pair of small n regions
We have observed previously JFET is biased in reverse biased to make a p n- junction with the help of insulated junction MOSFET device having the polarity positive or negative. (Malvino & Bates, 2007)
By doing this our MOSFET device electronic switch without having any bias and non-conducting obtaining voltage current devices with a very little amount, the P – channel and n- channel MOSFETs exists in two basic forms (S.W.AMOS, 1959)
Operational amplifier:
It is a device that consists all operations for amplification purpose and most commonly used in signal conditioning, filtering and to perform operations related to mathematics like addition, subtraction, integration as well as differentiation. It’s a voltage amplifying device having resistors and capacitors as feedback components in between their input and output these are mainly used at operating of amplifier with different configurations and perform wide variety of operations. Generally it’s a three terminal device containing very high impendence inputs which is inverting and non-inverting inputs followed by negative and positive signs. (Huijsing, 2011)
Figure 7:Basic Operational Amplifier
The terminal three indicated amplifier output port both sink and source can happen either a voltage or a current. In a linear operational amplifier, the output signal is the amplification factor, known as the amplifiers gain ( A ) multiplied by the value of the input signal and depending on the nature of these input and output signals, there can be four different classifications of operational amplifier gain.
The third terminal is the output port which is used a both source and sink like voltage or current. The output signal is nothing but the amplification factor in amplifiers which is known as gain where the input signal is multiplied by input value depending on these inputs and outputs there exists four amplifiers. (Nelson, 1995)
- Current = Current IN and Current OUT
- Trans conductance = Voltage IN and Current OUT
- Trans resistance =Current IN and Voltage OUT
- Voltage =Voltage IN and Voltage OUT
The output voltage signal from an Operational Amplifier is the difference between the signals being applied to its two individual inputs. Clearly saying that amplifiers signal is the difference between the two input signals in stage of signal amplifier id and bridge rectifier format, and by using class D amplifier will not become a huge cost saving one. In this amplifier the modifications and essentials were not justified in this area in order to use. Generally class D exists in the form of single IC having different output stages can find the chance of complexity inside the circuit, and can’t be discussed briefly, to plan and design class D amplifiers were not that much of effective. (Bugg, 1991)
Output filters:
In order to remove radiation produced at switching frequencies for amplifiers we use the output filter in order to improve the efficiency at external speaker cables. To allow some of the switching frequency energy from the from the inductance of loudspeaker and finally to pass to the ground giving rise to loses, since some applications of integrated low power does not contain output filters. Generally maximum class D filters contain second order LC filter in the middle of loudspeaker and amplifier. For better output we choose frequency having maximum flatness, especially when we use low impedance and separate loudspeakers class D amplifier is designed in such a way based on plausible assumptions, required the typical values inductor at 10µH-50µ which is too much greater than 1µH-2µH air-cored coils to provide stability at required loads in Class B amplifier. So mandatorily we use ferrite cored inductors and proper care must be taken at maximum point in order to eliminate saturation. (Temes & Mitra, 1973)
Why Filter is required in Class-D Amplifiers:
A Class-D amplifier output is a pulse-width-modulation (PWM) exchanged signal through duty cycle which is modulated with the audio signal. And the result contains the chosen audio range band frequencies and higher frequencies connected to the Pulse width modulation or converting frequency. The Class-D amplifiers, it is usually needed to pass this type signal over low pass filter to quotation audio content. Here low pass filter normally consists a series inductor also a capacitor connected to ground. In Class-D amplifiers a positive (OUT+) also the negative (OUT–) both outputs are continuously out of phase, by 50% duty cycle while zero Input applied. The result, for complete voltage at output is applied for the load in all times, and it cause to produces relatively very high current also great power dissipation at the load while no filter used. And efficiency of Class-D amplifier also reduced without any filter, and also quiescent current randomly increases. The Speaker impedance also includes inductance, however it is mostly resistive, where a LC filter is nearly purely sensitive. And LC filter through the cutoff frequency not as much of than a class-D switching frequency permits the switching current to stream through and filter instead for the load. A filter takes lower resistance compare the speaker, due to that it been decreases power dissipation, and increasing efficiency. (Bozic & Chance, 1998)
Class-D amplifiers usually usage low-pass type filter for attenuate the particular switching the noise in output waveform though fleeting the audio signal in to loudspeaker, The step in a Class-D amplifier used filter is a L-C low-pass filter. The cut off frequency in the filter chosen due to that filter will may take minimum effect at desired output of frequency range though attenuating switching the noise as far as possible.
Figure 8: Basic Class-D amplifier Low pass filter
A PWM switched signal containing duty signal which has been modulated with audio signal is the output of class D amplifier related to PWM it includes higher frequencies and band frequencies. In habitual class D amplifier it is mandatory that the signal must pass through low pass filter to obtain the desired audio content. In low pass filter a capacitor and filter is connected in series followed to ground, the positive and negative outputs are always out of phase followed by 50 % duty cycle even when zero input is applied. So at all times we can apply the full load output voltage which always gives rise to produce high voltage and current without using any filter, without filter we can reduce the efficiency and increase quiescent current and obviously inductance will be there at speaker impedance on the whole it is resistive and LC filter is completely reactive. Generally an LC filter having cut off frequency having small amount compared of class D switching frequency allows to flow the switching current to pass in filter instead of load because compared to speaker filter has a very low resistance which reduces power dissipation and increases frequency . To attenuate the switching noise at the level of output waveform when processing the audio signal towards the loud speaker the class D amplifiers uses low pass filter, but as per me to calculate and perform different components in class D the so many engineers does not the exact process. Since LC low pass filter is the heart of class D amplifier. In order to get minimal effect on desired output frequency generally we select a filter which is at corner frequency. Below figure shows the less attenuating and low switching noise. (Temes & Mitra, Modern filter theory and design , 1973)
The Low-Pass Filter:
Optimum value for the inductor filter is
L is RL/2πfc
Here fc= corner frequency in filter
RL means Load resistance
Depending upon speaker impedance the value of inductor changes
Here we need to calculate both capacitance and inductance independently. For that first need to find suitable inductor value and choose nearby value. Then better to calculate the capacitance value using inductor value.
Capacitance(C) = 1/ ((2πfc) 2 • L)
Here quality factor (Q) in a filter is the ratio at the center of frequency in the filter bandwidth.
Q (quality factor) = RL√(C/2L)
A high Q and low Q produces under and over damped curves. The filter range must be in 0.6 > Q > 0.8 to avoid under damped behavior, where equations explain about 0.7, which provides good performance and allows for impedance variation in the speakers ,here Q is filter change if the speaker impedance is changed without adjusting the filter component values, which results over and under damped results.
Component Selection
We can’t get the exact output by selecting the exact L-C filter but to select particular suitable components for the class D amplifier to eliminate loses and distortions at harmonic levels, the inductor’s DC current must be greater than or equal to maximum output current, the difference should not exceed between inductance and load current. We can even able to find the hysteresis loses depending upon the core material. So always the capacitor must be designed at multilayer polyester, polycarbonate or even with polypropylene film. In low pass filter it is better to avoid ceramic capacitors whereas in ceramic capacitors large amount of voltage change is there which finally causes distortion.
Topology Comparison – Linear vs. Class D
Here we discuss the differences of class D and other amplifiers. The first main difference is efficiency. Which is the exact reason to design class D amplifiers, the other amplifiers were very good in terms of performance, but having half inefficient rate, whereas a Class D amplifier is much more efficient, with values developed at practical designs.
Here we can clearly observe the difference between Class D and other amplifiers. Efficiency is the first and major difference in these which gave birth to class D amplifiers. In terms of performance they are all very linear having half of inefficient at efficient compared of class D amplifier. You can find below curves having difference in all terms.
Figure 9: Topology Comparison – Linear vs. Class D
Gain = Here gain is constant with different voltage levels at bus voltage but class D amplifier gain is proportional to bus voltage. Which is nothing but power supply rejection ration is very low and linear it is due to because they use feedback bus voltage variations, energy will always from towards load from the supply in full bridge and having two or more directional flow in half bridge to bus pumping criteria which helps to charge the capacitors of bus.
First, the FETs have a non-zero resistance even when they are turned hard on. This is typically in range 100 to 200mΩ, and can double as the device temperature increases from 0 to 150C, the latter being the usual maximum operating temperature. This resistance cause I
I2R losses. Second, the output devices have non-zero times for switching on and off. In the period when the FET is turning on or off, it has an intermediate value of resistance which again causes I
I2R losses. It is essential to minimize the stray inductance in the drain and source circuits as this not only extends the switching times but also cause voltage transients at turn-off that can overstress the transistor. Third, fly back pulses generated by an inductive load can cause conduction of the parasitic diodes have relatively long reverse recovery times and more current will flow than is necessary. To prevent this many Class-D designs have Schottky clamp diodes connected between the output line and supply rails. These turn on at a lower voltage than the parasitic FET diodes and deal with fly back pulses. They also have much faster recovery times.
Last, and perhaps most dangerous, is the phenomenon known as ‘shoot-through’. This somewhat opaque term refers to the situation when one FET has not stopped conducting before the other starts. This gives rise to an almost direct short between the supply rail, although very briefly, by a ‘dead-time’ circuit. The introduction of dead-time increase distortion, so only the minimum is applied; a 40nsec delay is sufficient to create more than 2% THD in 1 kHz sinewave
In FETs there is non- zero effect when they are ON existed in range between 100 to 200mΩ when the device temperature increases it can also be increased .This has IR loses and contains non- zero while switching points. So it is quite necessary to avoid these loses at drain and source and also generation of fly back pulses long reversed recovery time while conduction which requires more current. To eliminate these problems Class D amplifiers contains Schottky clamp diodes attached between the output line and supply rails. So due to this low voltage is generated when fly back pulses exists, and also faster recovery time very reasonable. This phenomenon is known as Shoot-Through which is quite dangerous in this condition we have to stop one FET while the other starts otherwise short will occur leaving the circuit dead and this increases distortion and can be applied a minimum input producing 1 kHz sinewave
The Class D Amplifier Advantage
- The supply is instantaneous to output current
- Better implementation that other amplifiers
- Large power dissipation even in most linear output stages
- Produces less heat, space and cost due to lower power dissipation
- Extends battery life. (Hood J. L., 1999)
Methodology:
The dissertation all about designing, simulating and making of digital audio amplifier using electronic discrete components. And the first thing need to design a suitable circuit diagram for requirements fulfil. Then need to design that circuit on the one of the electronic testing tool kind of Multisim, later need to check the running output by using simulating option in the Multisim, if there have any connection error then need to change and connect in proper way and also better to check after the connections in proper way better to look on oscilloscope toolbar the output which is getting on the screen is in right way which means fulfilling dissertation requirements or not otherwise need to change the connections again and also the components values which was used in circuit designing on Multisim and after getting proper output by running in the Multisim software, Need to convert the very same circuit on to breadboard for saving the money if they there have changes need to do before the PCB printing because some time few circuits will show proper output for the components which is used on Multisim but due to some components or company components will not give the same output which shown in the oscilloscope in Multisim.
For saving time and money better to produce the circuit on breadboard, once if the breadboard results also in the same way as Multisim then it’s good to move on Printed Circuit Board if not need to change the component values or components which is suitable based on project aims and applications.
Components which required for this Digital Class-D audio Amplifier designing as mentioned below.
Function generator
Operational amplifier (LM318D)
DC voltage Source (12V)
DC voltage source (15v)
MOSFET N channel (NTE4153NT1G)
MOSFET P channel (NTE4151PT1G)
For making Second Order low pass filter
Inductor
Capacitor
Load (on the simulation stage need a resistor)
In the hardware execution need a speaker
And also need connection wires in the breadboard execution stage
Processor: after drawing a suitable circuit diagram need to draw and execute in Multisim for the go to all apps in the lab computers and the click on faculty specific and select national instruments. In the national instruments go to circuit design suite 14.0 and select the Multisim 14.0 version option. From here onwards need to design the required circuit first need to select the required components for the circuit which is available in the commands. For this circuit first need an audio signal input for that need to select sinewave generator from place source and select AC POWER and place on the board which is showing in Multisim and then place a triangular wave generator also these two waves need to connect into Operational amplifier. For placing operational amplifier go to place Analog and then select suitable operational amplifier for the requirements fulfil which is LM318D and drag that component in to designing board. Here need to connect the sinewave generator in to pin 3 and triangular wave generator in to pin 2. The Vcc and Vee pin 7 and 4 respectively. These two need to connect one to +Ve voltage source and other one to –Ve voltage source. Here for checking instant out which came from the both the inputs and amplified signal in the operational amplifier can see if connect a oscilloscope next to the output pin of the operational amplifier. After connecting oscilloscope click on simulation button which is available on top of the design and observe the output it should be in Pulse Width Modulation otherwise need to go back and check the connections of inputs and values in sinewave generator and triangular wave generator.
After getting Pulse Width Modulation (PWM) the signal will go to switching stage. For designing switching stage here need two MOSFET. One need to belong to N channel and other one should be from P channel. After checking several transistors NTE4153NT1G and NTE4151PT1G these two suitable founded for required application. For placing these two MOSFET’s go to place a transistor command and select these two MOSFET’s from there and place next to Operational amplifier. The output signal from the operational amplifier directly connected to these two MOSFET’s, here also need voltage source which is about +15ve and -15ve respectively. For checking the switching output from the switching stage better to connect an oscilloscope next to switching stage and before the low pass filter. If the signal switched here successfully proceed to filter step otherwise need to check again the connections and components in the switching stage. In the filtering need to select proper values which is suitable for this circuit and input signal values. Usually the Class-D audio Amplifiers need Second order low-pass filter for avoiding noise from the switching stage, the reason behind choosing the Second order low-pass filter depends on the input signal and cut-off frequency. Here need to connect the RLC parts series connection with suitable values which is calculated below steps and simulate the entire circuit again with connecting an oscilloscope, and observe the final output in the very end oscilloscope. Based on the requirement and structure of the circuit here need to get a sinewave with little bit noise formation which is came in practically by the nature sounds or general sounds. But in the simulation for getting that noise we can connect an additional input very signal just before the operational amplifier and next to sinewave or with the sine. After executing successfully in the Multisim it’s time test on the breadboard the same circuit just before moving on to Printed Circuit Board (PCB). And here also need to connect the same connection as per the circuit but here need to give the sinewave generator and triangular wave generator physically which is available in electronic lab from the function generator. And also in the load place need to connect a speaker for hearing the output. In the input place we can give the microphone for giving input signal which is less than 20 kHz that’s the maximum audio signal can hear. And finally need to run the signal based on the input signals and observe the output in external connected oscilloscope of the circuit. After successfully completed breadboard simulation then need to print on PCB.
Simulation:
The first stage in the simulation to design the circuit, here designing the circuit step by step. And the first stage need to get the Pulse Width Modulation and the circuit for that as shown below.
Figure 10: Circuit diagram for PWM wave
Here in the above circuit diagram the the sine wave audio input as 1kHz given as one of the input to operational amplifier and triangular wave generator signal as 50 kHz and the voltage source also connected both Vcc and Vee respectively.
The value of the frequency in both the audio signal and the triangular signal as shown in the below boxes of screenshots
Figure 11: Triangular and sinewave frequencies
The output signal from the operational amplifier as like in below screenshot for the above circuit. And here the successful wave form of the Pulse width modulation for the Class-D audio amplifier.
Figure 12: Pulse Width Modulation signal from Operational Amplifier
And after getting Pulse Width Modulation proceed for the switching stage of the circuit here signal need to switch the signal for OFF and ON conditions. And the designed circuit for this as shown below,
Figure 13: Switching stage of Class-D audio Amplifier
And the output signal from the switching stage will switch the signal and the signal look as mention below,
Figure 14: Output signal from the Switching Stage
And then for clearing signal noise here need to use the second order low-pass filter and the filter design as shown in below.
LC=12πfc
20k=12×3.14LC
LC=1(2×3.14×20k)2
LC=63.32×10-12
C=1μF
L=63.32μH
And the filter design of the this circuit as shown in the below screenshot
Figure 15: Filter design for amplifier
As calculated the inductance and capacitance values as shown above equations arranged the inductance and capacitance and load then need to run the circuit again for the final time and click on the oscilloscope the output should be as shown in below screenshot,
Figure 16: Audio output signal
Conclusion and Future scope:
The design, simulating and making of the Class-D audio amplifier is partially succeed the pulse width modulation developing and switching signal in switching stage and making second order low-pass filter also developed in this dissertation. If execute on breadboard and the Printed Circuit Board and there have chance to get efficiency more than 90%.and also possible to get noise less signal in the final output. Due to lack of time and late start of my dissertation I’m unable to complete the practical work have chance to complete the practical work but don’t have exact components which I used in the designing the circuit in Multisim.
References
Bozic, S. M., & Chance, R. J. (1998). Digital filters and signal processing in electronic engineering : theory, applications, architecture, code. Chichester: Horwood Publishing,.
Bugg, D. V. (1991). Electronics : circuits, amplifiers and gates.
Duncan, B. (1996). High performance audio power amplifiers : for music performance and reproduction. Boston: Newnes.
Hood, J. L. (1997). Valve and transistor audio amplifiers.
Hood, J. L. (1999). Audio electronics. Oxford: Newnes.
Huijsing, J. H. (2011). Operational amplifiers : theory and design. (2nd Edition ed.). LONDON: springer.
Malvino, A. P., & Bates, D. J. (2007). Electronic principles (7th Edition ed.). LONDON: McGraw-Hill Higher Education,.
Nelson, J. C. (1995). Operation amplifier circuits : analysis and design. Boston: Butterworth-Heinemann.
S.W.AMOS. (1959). Princials Of Transistor Circuits (Eighth Edition ed.). oxford, United Kingdom: S.W.Amos .
Self, D. (2006). Audio Power Amplifier Design Handbook. Jordan Hill: Douglas Self.
Self, D. (2009). Handbook, Audio Power Amplifier Design (5th Edition ed.). Jordan Hill, United Kingdom: Elsevier Ltd.
Singmin, A. (2000). Practical audio amplifier circuit projects.
Slone, G. R. (1999). High-power audio amplifier construction manual : 50 to 500 watts for the audio perfectionist. New York: McGraw-Hill,.
Temes, G. C., & Mitra, S. K. (1973). Modern filter theory and design.
Temes, G. C., & Mitra, S. K. (1973). Modern filter theory and design . London: Wiley-Interscience.
Watkinson, J. (2001). The art of digital audio. (3rd Edition ed.). Oxford: Focal Press.
Leave a Reply