Power Amplifier

winetediousElectronics - Devices

Oct 7, 2013 (3 years and 8 months ago)

72 views


2
-
1

LABORATORY
2

POWER AMPLIFIER




OBJECTIVES

1.

To study Class B power amplifier circuits.

2.

To observe crossover distortion present in Class B power amplifiers.

3.

To simulate Class B and Class AB power amplifier circuits using
MicroCap

software.

4.

To design and test
DC biasing and frequency response of a
Class AB
audio power
amplifier.




INFORMATION


1. Power Amplifier Class B

Class B amplification involves using a dual voltage power supply along with two power
transistors, an NPN, and its complementary PNP device. Su
ch a circuit is shown in Figure
2.
1 and its operation could be explained as following:



In the absence of an input signal, neither transistor conducts; both transistors are
off.



On the positive half of the input cycle, once the input signal is greater than

0.7 V,
Q1 will turn on and current flows as shown in Figure
2.
1
-

a. Notice that the base
-
emitter voltage of Q1 causes Q2 to be held in the off state since Q2’s base
-
emitter
is reverse biased.




As the input signal swings into the negative half of its cycle

and exceeds 0.7V, Q2
is turned on and its base
-
emitter voltage reverse biases the base
-
emitter junction of
Q1, turning it off.
















a)
Positive half cycle operation


b) Class B output waveforms

Figure
2.
1
. Class B power amplifier operation




2
-
2

Typ
ical output waveforms for both Q1 and Q2 BJTs and a Class B amplifier output are
shown in Figure
2.
1
-
b.

The time required for the input signal to move from zero volts to +0.7 V or to
-
0.7 V is the
time during which conduction does not occur, consequently t
he output sits at zero volts for
this interval, producing what is called
crossover distortion
. Crossover distortion takes its
name from the dead
-
time distortion occurring when the input crosses over from
-
0.7 V to
+0.7 V or from +0.7 V to
-
0.7 V.


Class B

has a very low (
almost
zero) Quiescent Current, and hence low standing power
dissipation and optimum power efficiency. However it should be clear that in practice
Class B may suffer from problems when handling low
-
level signals.

In the absence of an
input

signal, a Class B power amplifier should have zero volts dc on the output terminal
with respect to ground, if the transistors are well matched. Often, they are not well
matched, so the student should be aware that it is quite possible to have a dc voltage

present at the output. Some output loads, such as speakers, may be damaged by dc. If such
loads are to be used, they must be capacitively coupled to the output in order to block the
dc.



2. Power Amplifier Class AB


Crossover distortion could be eliminat
ed in class AB power amplifiers by the addition of
the diode circuitry shown in Figure
2.
2a.
















a) Class AB circuit diagram



b) Class AB output waveforms

Figure
2.
2.
Class AB power amplifier circuit


Since the diodes in Figure
2.
2
-
a are on al
l the time, both Q1 and Q2 are held at the edge of
the conduction mode by the diode voltages (A

small but controlled Quiescent Current).
W
hen the input goes either positive or negative, very little voltage is required to put Q1 or
Q2 into full conduction.

Typical output waveforms for both Q1 and Q2 BJTs and a Class AB amplifier output are
shown in Figure
2.
2
-
b.






2
-
3


3. Transistors


You will be using the MJE800 NPN and the MJE700 PNP silicon Darlington pair power
transistors. These transistors are a set of co
mplimentary pair silicon power transistors.
Two individual transistors connected in a Darlington configuration in each package will
provide a very large short circuit current gain
β

which is the product of the two
β
’s of each
internal transistor. For the transistors used here the manufacturer guarantees a minimum
β

of 750. The transistor diagrams and package are shown in Figure
2.
A and the data sheets
are attached in the appendix
section of this manual.













a) MJE800 NPN and the MJE700 PNP diagrams


b) Package

Figure
2.
3.

MJE800 NPN and the MJE700 PNP diagrams and package


Note
: Two resistors and a diode are integrated internally in the transistor device’s package
and o
ne of the reasons for including these components is to prevent a thermal run
-
away
from occurring. These internal components are not shown on the circuit diagrams in
Figures
2.
1 and
2.
2 however they should be included in the device model in your circuit
si
mulation.




PRE
-
LABORATORY PREPARATI
ON

The lab preparation must be completed before coming to the lab. Show it to your TA for
checking and grading (out of 15) at the beginning of the lab and
get his/her signature.




1. Calculations

1.
The purpose of this ex
ercise is to design the output stage of an audio power amplifier

class AB
that could be used with one or more of the earlier circuits to complete a power
amplifier. In your design set the dual DC power supply to ± 6VDC. The amplifier should
deliver appr
oximately 500 mW of sinusoidal RMS audio power to an 8
Ω

load, over the
standard audio range of 20 Hz to 20 kHz. In the
laboratory,

you will use an 8.2
Ω

/5W load
resistor. It will make the lab a lot quieter! Include the basic power amplifier (Figure
2.
1)

and the diode compensated circuit (Figure
2.
4) in your pre
-
lab design and simulation. The
Figure
2.
4 circuit must be designed at the edge of the cut
-
off region. Since we are using a
Darlington pairs instead of single NPN and PNP transistors, the diode co
mpensation group
should contain three diodes instead of two, as it is shown in Figure
2.
4. The class AB



2
-
4

amplifiers have a small I
BIAS

such that the DC quiescent operating point is just into the start
of the conducting region. This will prevent a certain a
mount of cross over distortion.

2.
The class AB circuit must be designed at the edge of the cut
-
off region. For the circuit in
Figure
2.
4.calculate the values of the resistors R
1
=R
2

for a diode current of I
D
=5mA.















Figure
2.
4.

Real class AB po
wer amplifier circuit.


3
. For the circuit of Fig.
2.
4 calculate the

input power P
DC
, output power P
AC

and
the
efficiency


景f⁡渠楮灵n
獩s湡氠
噩渠V‴噰


(
畳楮g䕱畡瑩潮猠

ㄠ瑯1

5
)


䅳獵浥A



=

V
楮⸠
䕮瑥爠瑨r
牥獵s瑳⁩渠呡扬攠





2. MicroCap simulations

2.1.
The
MicroCap
-
9.0
(
Demo version)

has limited library and doesn’t provide Darlington
transistor models.
For these

simulations,

you will have to
u
se
MicroCap
-
9.0 (Professional
version)
available

only at SEB3108.

For
the
simulations use the
D
arlington transistor
model of TIP140
_FC

(to simulate
MJE800) and TIP145_FC (to simulate MJE700).



D
etermine the DC biasing volta
ges and currents with no ac signal for the
practical
class B
amplifier

circuit in Figure
2.
5
.
















Figure
2.
5.

Practical class B power amplifier




2
-
5

2.1.1.
Fill the Table
2.
1

with the
simulated

DC voltages when no AC input signal is
applied to the c
ircuit.

2.1.2. An input and output waveforms for a sinusoidal input signal Vin=4V
p
(peak) at
f=1kHz.

2.1.3. The output waveform for input voltage of Vin=8V
p
(peak) at 1 kHz. A
comparison with the voltages observed in the lab should be made. Watch for any
distortion occurring in the output waveform.


2.2.
Using calculated component values for resistors
R
1

and
R
2
.
determine the DC biasing
voltages and currents with no ac signal for the
class AB amplifier

circuit in Figure
2.
4
.
You should obtain the followin
g information through the
MicroCap

simulations:

2.2.1.
Fill up the T
able

2.
3

with the expected DC voltages when no AC input s
ignal is
applied to the circuit

2.2.2.
Print the
input and output waveforms for a sinusoidal input signal Vin = 4V
p
for
f =1 kHz.

2.2.3.
Obtain t
he frequency response of class AB amplifier from 10 Hz to 100 kHz.
Print the Bode plots of the voltage gain and
the
phase frequency response

of this amplifier

and bring these plots to the laboratory.

MicroCap simulations tips:



To provide a p
ower supply to the circuit use two “Battery” sources from the
MicroCap

library. Connect them as Vcc and Vee voltage sources with common
ground and set them to a 6VDC.



To obtain the values of all the bias currents and voltages on your schematic from
Analysi
s

menu choose the
Dynamic D
C

mode and click on
Node Voltages

and
Currents

icons on the toolbar.



For a sine wave signal source use a 1MHz Sinusoidal Source from the
Micro

Cap

library. Set the AC Amplitude to A=
4
(V) in the model description area of the sign
al
source.
Note

that A=4V corresponds to
V
p
=4V.



Run “TRANSIENT ANALYSIS” t
o obtain an input and out
put waveforms.



Run “AC ANALYSIS” t
o obtain
the gain and phase frequency response

plots for
this circuit for
frequency range from 10 Hz to 100 kHz.
Note
: Set

parameter P to
plot separate diagram for each curve.




EQUIPMENT

1.

Digital multimeter (Fluke 8010A, BK PRECISION 2831B).

2.

Function Generator Wavetek FG3B.

3.


Digital oscilloscope Tektronix TDS 210.

4.

MJE800 NPN and MJE700 PNP Darlington transistors.

5.

1N4148 diode
s


3.

6.

C=47

F


2; C=470uF


1.

7.

R=8.2


/ 2W.







2
-
6


PROCEDURE


1.

You are provided with two heat sinks, which should be attached to the transistors
during the lab exercise.
The
heat sink
supposed to be
electrically
insulated from

the collector of the transis
tor
, however it is always recommended to avoid any
contact of the heat sinks to the ground or to each other
. Occasionally check the
temperature of the heat sink, if you cannot keep your finger of the heat sink for
more than twenty seconds the transistors
may be too hot. Shut the power off and
check your circuit.


2.

Connect the
class B

power amplifier shown in Figure
2.
5 using
MJE800 NPN and
MJE700 PNP Darlington transistors

instead of single BJTs.

Use the R
L
= 8.2
Ω

resistor to
replace

the loudspeaker’s load.


3.

Use a dual voltage Power Supply and connect its POS terminal as Vcc, NEG
terminal as Vee and COM terminal as a common ground. Set the power supply
voltage to 6V DC. Measure the DC quiescent point values. Com
pare the voltages
and currents from simulation with the experimental data in a Table
2.
1. If your
results are significantly different (more than 15%) from your simulated values, try
to find out and eliminate the reason for that discrepancy.

.


Q1

Q2


V
CE

[V]

V
BE

[V]

I
C

[A]

V
CE

[V]

V
BE

[V]

I
C

[A]

Simulation







Experiment







Table
2.
1.
Class B power amplifier DC biasing


4.

Once you are satisfied that your circuit is biased correctly, then connect the signal
generator to the input. Set the signal gene
rator to a frequency of 1 kHz. For the
input signal level of Vin =
3V
rms

(~
4Vp
)

sketch the output voltage across the 8.2
Ω

load on top of your MicroCap simulation plot. Compare the simulated and
experimental waveforms and explain the

differences if any.


5.

Increase the input sinusoidal voltage until you notice a clipping in the output
voltage.
Record this value and compare with DC power supply voltages.

For these
readings you can use the
BK Precision

meter to measure the AC input current (
it
measures the RM
S value), measure the input voltage after the
digital

meter (scope)
as it is shown in Figure
2.
6
.


6.

For the
input signal
Vin= 2

V
rms

and Vin=
1.8
V
rms

calculate the input AC power P
in
,
the output AC power P
o
, the DC input power from the DC supply P
DC
. Also
ca
lculate the AC voltage gain A
V

[dB] (Equation (
2.
1)), the AC power gain [dB]
(Equation (
2.
2.)) and the amplifier efficiency

(Equation (
2.
5
)) of the class B
power amplifier.





2
-
7














Figure
2.
6
.
Class B power amplifier measurements
.





Equation (
2.
1)






Equation (
2.
2)







Equation (
2.
3)







Equation (
2.
4)






Equation (
2.
5
)


Record your measurements and calculati
ons in Table
2.
2. Determine if your amplifier
is capable of delivering 500 mW of audio power without distortion. If your circuit can
not deliver this power, do not lay the sole blame on the DC power supply, the
maximum current it can deliver is 200 mA.


A
C input

measurements

AC output

measurements

DC input

measurements

Calculations

V
in

[V]

I
in

[A]

P
in

[W]

V
o
(p)

[V]

P
o
AC

[W]

V
DC

[V]

P
DC

[W]

A
V

[dB]

A
P

[dB]























Table
2.
2.

Class B power amplifier measurements.


7.

Connect the class AB power

amplifier in Figure
2.
4. Use the calculated values of
R
1

and R
2
. Repeat the DC biasing measurements from point 2 and collect all data
in Table
2.
3.



2
-
8


Q1

Q2


V
CE

[V]

V
BE

[V]

I
C

[A]

V
CE

[V]

V
BE

[V]

I
C

[A]

Simulation







Experiment







Table
2.
3.
Cl
ass AB power amplifier DC biasing


8.

Repeat all measurements from points 3 and 4 and collect all data in Table
2.
4.

Compare the results with the pre
-
lab calculations.

AC input

measurements

AC output

measurements

DC input

measurements

Calculations

V
in

[V]

I
i
n

[A]

P
in

[W]

V
o
(p)

[V]

P
o
AC

[W]

V
DC

[V]

P
DC

[W]

A
V

[dB]

A
P

[dB]



3










2










ㄮ1










Table
2.
4.

Class AB power amplifier AC measurements.


9.

Determine the frequency response of the AC sinusoidal voltage gain of the
compensated

amplifier
over the range of 20 Hz to 30 kHz.
















Table
2.
5
.

Frequency response of a class AB power amplifier.


10.

Plot obtained voltage gain and phase data on top of your simulated Bode plots and
compare the results.




REPORT

Your Lab report is due one wee
k later. Please submit it to your TA in the beginning of
your next lab session.

Note
: You must copy/print the Signature and Marking Sheet from your manual
before coming to the lab session.

f

孈[]

V
in

[V]

V
o

[V]


de杝

A
v
[dB]

20





50





100





200





500





800





1k





5k





10k





20k





30k







2
-
9



SIGNATURE AND MARKIN
G SHEET


LAB
2


To be completed by TA durin
g your lab session


Student Name:____________________

TA Name:___________________

Student # : _____________________


Check
boxes

Task

Max.

Marks

Granted

Marks

TA

Signature




Pre
-
lab completed

15






Class B Amplifier Test completed

2
0






Class AB A
mplifier Test completed

2
0






Overall Report Preparation

4
5






TOTAL MARKS

100