1
00:00:00,000 --> 00:00:02,000
Good morning,
all.

2
00:00:02,000 --> 00:00:08,000
Good morning.
I hope you guys did not spend

3
00:00:08,000 --> 00:00:14,000
all of last night celebrating
the Red Sox victory,
And today we will take the next

4
00:00:14,000 --> 00:00:19,000
but there is one more tonight.
OK.

5
00:00:19,000 --> 00:00:24,000
Let's see.
I trust the quiz went OK.

6
00:00:24,000 --> 00:00:32,000
What I will do today is take
off from where we left off on

7
00:00:32,000 --> 00:00:38,000
Tuesday.
And continue our discussion of

8
00:00:38,000 --> 00:00:44,000
the large signal and small
signal analysis of our

9
00:00:44,000 --> 00:00:49,000
amplifier.
Today the focus will be on

10
00:00:49,000 --> 00:00:53,000
"Small Signal Analysis".

11
00:01:02,000 --> 00:01:09,000
So let me start by reviewing
some of the material.

12
00:01:09,000 --> 00:01:16,000
And, as you know,
our MOSFET amplifier looks like

13
00:01:16,000 --> 00:01:18,000
this.

14
00:01:30,000 --> 00:01:33,000
One of the things you will
notice in circuits,

15
00:01:33,000 --> 00:01:37,000
as I have been mentioning all
along in this course,

16
00:01:37,000 --> 00:01:42,000
is that certain kinds of
patterns keep repeating time and

17
00:01:42,000 --> 00:01:45,000
time again.
And this is one such pattern.

18
00:01:45,000 --> 00:01:50,000
A three terminal device like
the MOSFET with an input and the

19
00:01:50,000 --> 00:01:55,000
drain to source port connected
to RL and VS in series in the

20
00:01:55,000 --> 00:02:00,000
following manner,
this is a very common pattern.

21
00:02:00,000 --> 00:02:02,000
There are several other common
patterns.

22
00:02:02,000 --> 00:02:05,000
The voltage divider is a common
pattern.

23
00:02:05,000 --> 00:02:08,000
We keep running into that again
and again and again.

24
00:02:08,000 --> 00:02:12,000
The Thevenin form,
a voltage source in series with

25
00:02:12,000 --> 00:02:14,000
the resistor is another very
common form.

26
00:02:14,000 --> 00:02:18,000
The Norton equivalent form,
which is a current source in

27
00:02:18,000 --> 00:02:21,000
parallel with a resistor is also
very common.

28
00:02:21,000 --> 00:02:25,000
And it behooves all of us to be
very familiar with the analyses

29
00:02:25,000 --> 00:02:30,000
of these things.
Voltage dividers in particular

30
00:02:30,000 --> 00:02:34,000
are just so common that you need
to be able to look at it and

31
00:02:34,000 --> 00:02:38,000
boom, be able to write down the
expression for voltage dividers.

32
00:02:38,000 --> 00:02:42,000
I would also encourage you to
go and look at current dividers.

33
00:02:42,000 --> 00:02:46,000
When you have two resistors in
parallel and you have some

34
00:02:46,000 --> 00:02:49,000
current flowing into the
resistors to find out the

35
00:02:49,000 --> 00:02:53,000
current in one branch versus the
other very quickly.

36
00:02:53,000 --> 00:02:57,000
The expression is very
analogous to the voltage divider

37
00:02:57,000 --> 00:03:01,000
expression.
And some of these very common

38
00:03:01,000 --> 00:03:06,000
patterns are highlighted in the
summary pages in the course

39
00:03:06,000 --> 00:03:11,000
notes, so it is good to keep
track of those and be extremely

40
00:03:11,000 --> 00:03:16,000
familiar with those patterns to
the point where if you see it

41
00:03:16,000 --> 00:03:21,000
you should be able to jump up
and shout out the answer just by

42
00:03:21,000 --> 00:03:24,000
looking at it without having to
do any math.

43
00:03:24,000 --> 00:03:29,000
So here was an amplifier.
And then we noticed that when

44
00:03:29,000 --> 00:03:33,000
the MOSFET was in saturation it
behaved like a current source.

45
00:03:33,000 --> 00:03:37,000
And this circuit would give us
amplification while the MOSFET

46
00:03:37,000 --> 00:03:40,000
was in saturation.
So we agreed to adhere to the

47
00:03:40,000 --> 00:03:43,000
saturation discipline which
simply said that I was going to

48
00:03:43,000 --> 00:03:47,000
use my circuit in a way that the
MOSFET would always remain in

49
00:03:47,000 --> 00:03:50,000
saturation in building things
like amplifiers and so on.

50
00:03:50,000 --> 00:03:54,000
And by doing that throughout
the analysis I could make the

51
00:03:54,000 --> 00:03:57,000
assumption that the MOSFET was
in saturation.

52
00:03:57,000 --> 00:04:01,000
I didn't have to go through --
Analysis became easier.

53
00:04:01,000 --> 00:04:05,000
I didn't have to figure out
now, what region is the MOSFET

54
00:04:05,000 --> 00:04:07,000
in?
Well, because of my discipline

55
00:04:07,000 --> 00:04:10,000
it is always going to be in
saturation.

56
00:04:10,000 --> 00:04:13,000
But in turn what we had to do
was conduct a large signal

57
00:04:13,000 --> 00:04:15,000
analysis.

58
00:04:22,000 --> 00:04:25,000
Again, in follow on courses you
will be given circuits like

59
00:04:25,000 --> 00:04:28,000
this.
In fact, this very circuit with

60
00:04:28,000 --> 00:04:31,000
a very high likelihood.
And you will be looking at more

61
00:04:31,000 --> 00:04:33,000
complicated models of the
MOSFET.

62
00:04:33,000 --> 00:04:36,000
Or you will be given the MOSFET
like this and,

63
00:04:36,000 --> 00:04:40,000
let's say in that course the
designers do not adhere to the

64
00:04:40,000 --> 00:04:43,000
saturation discipline,
in which case you have to first

65
00:04:43,000 --> 00:04:46,000
figure out is my MOSFET in its
triode region or in the

66
00:04:46,000 --> 00:04:49,000
saturation region?
And depending on the region it

67
00:04:49,000 --> 00:04:52,000
is in you have to apply
different equations.

68
00:04:52,000 --> 00:04:57,000
So it is one step more
complicated than in 002.

69
00:04:57,000 --> 00:05:00,000
In 002 we simplified our lives
by following a discipline.

70
00:05:00,000 --> 00:05:04,000
And let me tell you that
following a discipline is quite

71
00:05:04,000 --> 00:05:06,000
OK.
When it simplifies our lives

72
00:05:06,000 --> 00:05:10,000
and we can do good things with
it, it is quite OK to do that.

73
00:05:10,000 --> 00:05:12,000
We are not wimps or anything
like that.

74
00:05:12,000 --> 00:05:16,000
It is quite OK to have a
discipline and agree that we are

75
00:05:16,000 --> 00:05:19,000
going to play in this region of
the playground and build

76
00:05:19,000 --> 00:05:22,000
circuits in that manner.
By doing so,

77
00:05:22,000 --> 00:05:25,000
we could assume the MOSFET was
in saturation all the time.

78
00:05:25,000 --> 00:05:30,000
And analysis simply used a
current source model.

79
00:05:30,000 --> 00:05:33,000
By the same token,
what becomes important is to

80
00:05:33,000 --> 00:05:38,000
figure out what are the
boundaries of valid operation of

81
00:05:38,000 --> 00:05:43,000
the MOSFET in saturation?
To do that we conducted a large

82
00:05:43,000 --> 00:05:46,000
signal analysis.
And it had two components to

83
00:05:46,000 --> 00:05:49,000
it.
One of course was to figure out

84
00:05:49,000 --> 00:05:52,000
the output versus input
response.

85
00:05:52,000 --> 00:05:56,000
And what this usually does is
that it does a nonlinear

86
00:05:56,000 --> 00:06:02,000
analysis of this circuit.
If it is a linear circuit it is

87
00:06:02,000 --> 00:06:05,000
a linear analysis.
And figures out what the values

88
00:06:05,000 --> 00:06:09,000
of the various voltages and
currents are in the circuit as a

89
00:06:09,000 --> 00:06:12,000
function of the applied inputs
and chosen parameters.

90
00:06:12,000 --> 00:06:16,000
And the second step we said was
to figure out valid operating

91
00:06:16,000 --> 00:06:18,000
ranges --

92
00:06:27,000 --> 00:06:33,000
-- for input and corresponding
ranges for the other dependent

93
00:06:33,000 --> 00:06:38,000
parameters such as VO.
You could also find out the

94
00:06:38,000 --> 00:06:44,000
corresponding operating range
for the current IDS and so on.

95
00:06:44,000 --> 00:06:49,000
So by doing this you could
first analyze the circuit,

96
00:06:49,000 --> 00:06:55,000
find out the "bias" parameters,
find out the values of VI and

97
00:06:55,000 --> 00:07:00,000
VO and so on.
And then you could say all

98
00:07:00,000 --> 00:07:04,000
right, provided,
as long as VI stays within

99
00:07:04,000 --> 00:07:10,000
these bounds my assumption that
this is in saturation will hold

100
00:07:10,000 --> 00:07:15,000
and everything will be fine.
The reading for this is Chapter

101
00:07:18,000 --> 00:07:22,000
step and revisit small signal
analysis.

102
00:07:22,000 --> 00:07:27,000
In the demo that I showed you
at the end of last lecture,

103
00:07:27,000 --> 00:07:32,000
I showed you an input
triangular wave.

104
00:07:32,000 --> 00:07:36,000
And the input triangular wave
gave rise to an output.

105
00:07:36,000 --> 00:07:39,000
And we noticed that we did have
amplification,

106
00:07:39,000 --> 00:07:43,000
I had a small input and a much
bigger output.

107
00:07:43,000 --> 00:07:47,000
I did have amplification when
the MOSFET was in saturation but

108
00:07:47,000 --> 00:07:52,000
it was highly nonlinear.
The input was a triangular wave

109
00:07:52,000 --> 00:07:56,000
and the output was some funny,
it kind of looked like a

110
00:07:56,000 --> 00:08:01,000
sinusoid whose extremities had
been whacked down and kind of

111
00:08:01,000 --> 00:08:05,000
flattened.
And its upward going peak had

112
00:08:05,000 --> 00:08:08,000
been shrunk.
So it was a kind of weird

113
00:08:08,000 --> 00:08:12,000
nonlinear behavior.
I will show that to you again

114
00:08:12,000 --> 00:08:15,000
later on.
And so it amplified but it was

115
00:08:15,000 --> 00:08:18,000
nonlinear.
And remember our goal of two

116
00:08:18,000 --> 00:08:21,000
weeks ago?
We set out to build a linear

117
00:08:21,000 --> 00:08:25,000
amplifier.
So today we will walk down that

118
00:08:25,000 --> 00:08:30,000
path and talk about building a
linear amplifier.

119
00:08:30,000 --> 00:08:35,000
So to very quickly revisit the
input versus output

120
00:08:35,000 --> 00:08:39,000
characteristic,
VI versus VO,

121
00:08:39,000 --> 00:08:45,000
this is VT and this is VS,
this is what things looked

122
00:08:45,000 --> 00:08:49,000
like.
Also to quickly review the

123
00:08:49,000 --> 00:08:53,000
valid ranges,
until some point here the

124
00:08:53,000 --> 00:09:00,000
amplifier was in saturation,
the MOSFET was in saturation

125
00:09:00,000 --> 00:09:06,000
and somewhere here I had VO
being equal to VI minus a

126
00:09:06,000 --> 00:09:12,000
threshold drop.
At that point the MOSFET went

127
00:09:12,000 --> 00:09:18,000
into its triode region and I no
longer was following the

128
00:09:18,000 --> 00:09:23,000
saturation discipline.
So therefore this is my valid

129
00:09:23,000 --> 00:09:28,000
region of operation.
We also know that the output

130
00:09:28,000 --> 00:09:35,000
was given by VS minus K (VI-VT)
all squared RL over 2.

131
00:09:35,000 --> 00:09:38,000
Again assuming the MOSFET is in
saturation.

132
00:09:38,000 --> 00:09:43,000
It is very important to keep
stating this because this is

133
00:09:43,000 --> 00:09:49,000
true only when the MOSFET is in
saturation, when I am following

134
00:09:49,000 --> 00:09:53,000
the discipline.
Notice that this is a nonlinear

135
00:09:53,000 --> 00:09:56,000
relationship.
So VO depends on some funny

136
00:09:56,000 --> 00:10:02,000
square law dependence on VI.
The key here is how do we go

137
00:10:02,000 --> 00:10:08,000
about building our amplifier?
Take a look at this point here.

138
00:10:08,000 --> 00:10:13,000
At this point here let's say I
have a VI input.

139
00:10:13,000 --> 00:10:17,000
Corresponding output is VO.
Focus is this point.

140
00:10:17,000 --> 00:10:22,000
And left to itself this was a
nonlinear curve.

141
00:10:22,000 --> 00:10:28,000
Remember the trick that we used
in our nonlinear Expo Dweeb

142
00:10:28,000 --> 00:10:31,000
example?
We used the Zen Method.

143
00:10:31,000 --> 00:10:34,000
Remember the Zen Method?
We said look,

144
00:10:34,000 --> 00:10:38,000
this is nonlinear,
but if you can focus your mind

145
00:10:38,000 --> 00:10:42,000
on this little piece of the
curve here this looks more or

146
00:10:42,000 --> 00:10:46,000
less linear.
If I look at a small itty-bitty

147
00:10:46,000 --> 00:10:49,000
portion of the curve and I do
the Zen thing,

148
00:10:49,000 --> 00:10:53,000
and kind of zoom in on here.
This looked more or less

149
00:10:53,000 --> 00:10:56,000
linear.
This means that if I could work

150
00:10:56,000 --> 00:11:01,000
with very small signals and
apply the signal in a way that I

151
00:11:01,000 --> 00:11:05,000
also had a DC offset of some
sort.

152
00:11:05,000 --> 00:11:08,000
Then I would be in a region of
the curve, I would be

153
00:11:08,000 --> 00:11:12,000
delineating a small region of
the curve which would be more or

154
00:11:12,000 --> 00:11:15,000
less linear.
This was a small signal trick.

155
00:11:15,000 --> 00:11:19,000
And what we will do here is
simply revisit the small signal

156
00:11:19,000 --> 00:11:21,000
model.
Most of what I am going to do

157
00:11:21,000 --> 00:11:25,000
from here on will be more or
less a repeat of what you saw

158
00:11:25,000 --> 00:11:29,000
for the light emitting expo
dweeb.

159
00:11:29,000 --> 00:11:32,000
Just that here I have a three
terminal device,

160
00:11:32,000 --> 00:11:35,000
with a little bit more
complication.

161
00:11:35,000 --> 00:11:40,000
The equation is different.
I don't have to resort to a

162
00:11:40,000 --> 00:11:44,000
Taylor series expansion.
I will just do a complete

163
00:11:44,000 --> 00:11:50,000
expansion of this expression and
develop the small signal values

164
00:11:50,000 --> 00:11:53,000
for you.
Recall the small signal model.

165
00:11:53,000 --> 00:12:01,000
It had the following steps.
The first step will operate at

166
00:12:01,000 --> 00:12:08,000
some bias point,
VI, VO, and of course some

167
00:12:08,000 --> 00:12:14,000
corresponding point IDS.
This is Page 3.

168
00:12:14,000 --> 00:12:24,000
And then superimpose a small
signal VI on top of the big fat

169
00:12:24,000 --> 00:12:30,000
bias.
Remember the "boost"?

170
00:12:30,000 --> 00:12:32,000
So VI is the boost.
Boom.

171
00:12:32,000 --> 00:12:38,000
And above VI,
I have small signal VI that I

172
00:12:38,000 --> 00:12:42,000
apply.
And our claim is that response

173
00:12:42,000 --> 00:12:49,000
of the amplifier to VI is
approximately linear.

174
00:13:04,000 --> 00:13:09,000
The key trick with this is that
for my small signal model here,

175
00:13:09,000 --> 00:13:12,000
this is Page 3 here,
and Page 2.

176
00:13:12,000 --> 00:13:16,000
The key trick here is that with
the small signal model,

177
00:13:16,000 --> 00:13:20,000
I operate my amplifier at some
operating point,

178
00:13:20,000 --> 00:13:24,000
VO, VI.
I superimpose a small signal VI

179
00:13:24,000 --> 00:13:27,000
on top of small VI on top of big
VI.

180
00:13:27,000 --> 00:13:32,000
And then I claim that the
response to VI is approximately

181
00:13:32,000 --> 00:13:36,000
linear.
And let me just embellish that

182
00:13:36,000 --> 00:13:38,000
curve a little bit more.

183
00:13:43,000 --> 00:13:46,000
Notice that in this situation
this was my VI,

184
00:13:46,000 --> 00:13:51,000
which is my bias voltage,
this is VO, which is the output

185
00:13:51,000 --> 00:13:56,000
bias, and of course not shown on
this graph is the output

186
00:13:56,000 --> 00:14:03,000
operating current which is IDS.
One nice way of thinking about

187
00:14:03,000 --> 00:14:10,000
this is to redraw this and think
that your coordinate axes have

188
00:14:10,000 --> 00:14:14,000
kind of shifted in the following
manner.

189
00:14:14,000 --> 00:14:18,000
This is VI.
This is also on your Page 3.

190
00:14:18,000 --> 00:14:23,000
This is VT.
Remember this was the operating

191
00:14:23,000 --> 00:14:27,000
point, VO and VI.
And notice that we were

192
00:14:27,000 --> 00:14:35,000
operating in this small regime
of our transfer curve here.

193
00:14:35,000 --> 00:14:39,000
And in effect what we are
saying is that I am going to

194
00:14:39,000 --> 00:14:45,000
apply small variations about VI
and call those variations delta

195
00:14:45,000 --> 00:14:49,000
VI or small VI.
And the resulting variations

196
00:14:49,000 --> 00:14:55,000
are going to look like delta VO.
Also referred to as small V,

197
00:14:55,000 --> 00:14:58,000
small O.
So I will have small variations

198
00:14:58,000 --> 00:15:01,000
here.
And they give rise to

199
00:15:01,000 --> 00:15:04,000
corresponding small variations
there.

200
00:15:04,000 --> 00:15:09,000
One way to view this is as if
we are working with a new

201
00:15:09,000 --> 00:15:12,000
coordinate system.
Another way to view this is

202
00:15:12,000 --> 00:15:17,000
that so the capital VI and
capital VO correspond to my VI

203
00:15:17,000 --> 00:15:22,000
and VO as the total voltages in
my circuit, but at this bias

204
00:15:22,000 --> 00:15:26,000
point I can think of another
coordinate system here with

205
00:15:26,000 --> 00:15:32,000
small VI and VO out there.
And for small changes to VI,

206
00:15:32,000 --> 00:15:37,000
I can figure out the
corresponding small changes to

207
00:15:37,000 --> 00:15:40,000
VO.
Just that all the analysis I

208
00:15:40,000 --> 00:15:43,000
perform here is going to be
linear.

209
00:15:43,000 --> 00:15:48,000
And I will prove it to you in a
couple of different ways in the

210
00:15:48,000 --> 00:15:53,000
next few seconds.
When I am doing small signal

211
00:15:53,000 --> 00:16:00,000
analysis I am operating here in
this regime at some bias point.

212
00:16:00,000 --> 00:16:05,000
You have also seen this before.
How do I get a bias?

213
00:16:05,000 --> 00:16:10,000
This is my amplifier RL and VS.
This is Page 4.

214
00:16:10,000 --> 00:16:14,000
VO.
The way I get a bias is I apply

215
00:16:14,000 --> 00:16:20,000
some DC voltage VI and
superimpose on top of that my

216
00:16:20,000 --> 00:16:25,000
small signal small VI.
This is my DC bias that has

217
00:16:25,000 --> 00:16:32,000
boosted up the signal to an
interesting value.

218
00:16:32,000 --> 00:16:37,000
And because of that what I can
get is by varying VI as a small

219
00:16:37,000 --> 00:16:42,000
signal with a very small
amplitude, I am going to get a

220
00:16:42,000 --> 00:16:47,000
linear response here.
And I can draw that for you as

221
00:16:47,000 --> 00:16:48,000
well.

222
00:16:54,000 --> 00:17:00,000
This is my bias point here.
And if I vary my signal like so

223
00:17:00,000 --> 00:17:05,000
then my output should look like
this.

224
00:17:05,000 --> 00:17:09,000
This is point VI,
this is point VO,

225
00:17:09,000 --> 00:17:17,000
and this is my small signal VI
and this is my small signal VO

226
00:17:17,000 --> 00:17:23,000
and this is capital VO.
So this small thing here is VI.

227
00:17:23,000 --> 00:17:30,000
I would like to show you a
little demo.

228
00:17:30,000 --> 00:17:34,000
I will start with the same demo
I showed you the last time.

229
00:17:34,000 --> 00:17:39,000
I showed you the amplifier.
In the demo I am going to apply

230
00:17:39,000 --> 00:17:42,000
a triangular wave.
And initially I start with a

231
00:17:42,000 --> 00:17:45,000
large signal.
And you will see that the

232
00:17:45,000 --> 00:17:50,000
output looks really corny,
is going to look something like

233
00:17:50,000 --> 00:17:52,000
this.
That's large signal response.

234
00:17:52,000 --> 00:17:57,000
And then I will begin playing
with the input making it

235
00:17:57,000 --> 00:18:02,000
smaller, and you can see how it
looks yourselves.

236
00:18:02,000 --> 00:18:06,000
There you go.
So this is where I stopped the

237
00:18:06,000 --> 00:18:10,000
last time.
The last lecture I applied this

238
00:18:10,000 --> 00:18:16,000
input, time is going to the
right, and the purple curve in

239
00:18:16,000 --> 00:18:22,000
the background is the output.
It looks much more like a

240
00:18:22,000 --> 00:18:26,000
sinusoid with some flattening of
its tips.

241
00:18:26,000 --> 00:18:32,000
Nothing like an interesting
triangular wave.

242
00:18:32,000 --> 00:18:37,000
What I will do next is that let
me make sure I have enough of a

243
00:18:37,000 --> 00:18:41,000
boost here, enough of a DC
voltage so that I am operating

244
00:18:41,000 --> 00:18:45,000
at some point here.
I believe I already have that.

245
00:18:45,000 --> 00:18:49,000
Notice that I can shift up the
triangular wave input,

246
00:18:49,000 --> 00:18:53,000
or I can shift it down.
So let me bias it here.

247
00:18:53,000 --> 00:18:57,000
I have chosen a VI that's
about, I forget how many volts

248
00:18:57,000 --> 00:19:03,000
per division it is,
but I have chosen some VI here.

249
00:19:03,000 --> 00:19:06,000
And I biased it such that this
is the input.

250
00:19:06,000 --> 00:19:09,000
You get a nonlinear response.
It is amplified.

251
00:19:09,000 --> 00:19:12,000
It is much bigger.
What I will do next is make VI

252
00:19:12,000 --> 00:19:14,000
that I apply smaller and
smaller.

253
00:19:14,000 --> 00:19:17,000
I have already done the
boosting.

254
00:19:17,000 --> 00:19:20,000
Boom, that's a boost.
So I have boosted up your VI

255
00:19:20,000 --> 00:19:23,000
already.
Next is I am going to shrink

256
00:19:23,000 --> 00:19:27,000
it, and hopefully you will see
that if all that I am saying is

257
00:19:27,000 --> 00:19:32,000
truthful here you will see a
triangular response.

258
00:19:32,000 --> 00:19:35,000
Let's go try it out.
Watch the yellow.

259
00:19:35,000 --> 00:19:42,000
I am going to shrink the yellow
and make it smaller and smaller.

260
00:19:42,000 --> 00:19:47,000
There you go.
It is great when nature works

261
00:19:47,000 --> 00:19:52,000
like you expect it to.
I have never seen a triangular

262
00:19:52,000 --> 00:19:57,000
wave looks so pretty in my life.
It is awesome.

263
00:19:57,000 --> 00:20:03,000
Look at this.
Here is a tiny triangular wave.

264
00:20:03,000 --> 00:20:08,000
And the output is also a
triangular wave but it is much

265
00:20:08,000 --> 00:20:10,000
more linear.
Yes.

266
00:20:10,000 --> 00:20:12,000
Question?
What's that?

267
00:20:12,000 --> 00:20:18,000
The question is that the output
here is only as big as the input

268
00:20:18,000 --> 00:20:22,000
used to be before.
That's a good question.

269
00:20:22,000 --> 00:20:27,000
What I have done here is I am
showing you a laboratory

270
00:20:27,000 --> 00:20:31,000
experiment.
And let's assume that this

271
00:20:31,000 --> 00:20:35,000
input is the input I am getting
from some sensor in the field.

272
00:20:35,000 --> 00:20:38,000
Assume that this is my input,
not what I had before.

273
00:20:38,000 --> 00:20:41,000
Assume that this is my input to
begin with and this is the

274
00:20:41,000 --> 00:20:44,000
amplified output.
What I can also do is I can

275
00:20:44,000 --> 00:20:47,000
also change the bias.
And we will see this at the end

276
00:20:47,000 --> 00:20:49,000
of the lecture,
in the last ten minutes of

277
00:20:49,000 --> 00:20:51,000
lecture.
How do you select a bias point?

278
00:20:51,000 --> 00:20:55,000
By changing your bias point you
can change the properties of an

279
00:20:55,000 --> 00:21:00,000
amplifier to give you a preview
of upcoming attractions.

280
00:21:00,000 --> 00:21:02,000
Let me ask you,
what do you think should happen

281
00:21:02,000 --> 00:21:06,000
if I change the bias point?
I have not shown you the math

282
00:21:06,000 --> 00:21:09,000
yet, so intuitively what do you
think should happen?

283
00:21:09,000 --> 00:21:13,000
If I increase the bias what do
you think is going to happen?

284
00:21:13,000 --> 00:21:14,000
Yes.
Good insight.

285
00:21:14,000 --> 00:21:17,000
Higher bias will be more
amplification.

286
00:21:17,000 --> 00:21:20,000
Let's see if our friend is
correct.

287
00:21:33,000 --> 00:21:36,000
Let me set a higher bias.

288
00:21:44,000 --> 00:21:45,000
Not necessarily,
I guess.

289
00:21:45,000 --> 00:21:47,000
You're actually right,
by the way.

290
00:21:47,000 --> 00:21:50,000
I am playing a trick on
everybody here.

291
00:22:02,000 --> 00:22:05,000
As I change my input bias.
Notice that under certain

292
00:22:05,000 --> 00:22:10,000
conditions my output becomes
smaller and gets more distorted.

293
00:22:10,000 --> 00:22:14,000
Under other conditions what is
going to happen to my output is

294
00:22:14,000 --> 00:22:19,000
that it is becoming smaller and
is going to get distorted again.

295
00:22:19,000 --> 00:22:23,000
So there are a bunch of funny
effects happening that reflect

296
00:22:23,000 --> 00:22:26,000
on the bias point,
but for an appropriate choice

297
00:22:26,000 --> 00:22:31,000
of bias point as I increase the
bias the amplification should

298
00:22:31,000 --> 00:22:34,000
increase.
And I will show you that in a

299
00:22:34,000 --> 00:22:36,000
few minutes.
But it is a complicated

300
00:22:36,000 --> 00:22:38,000
relationship.
Yes.

301
00:22:44,000 --> 00:22:47,000
This is finally getting fun.
Here is the question.

302
00:22:47,000 --> 00:22:50,000
Professor Agarwal,
we love your song and dance,

303
00:22:50,000 --> 00:22:53,000
but if you really want to get a
high signal at the output and

304
00:22:53,000 --> 00:22:58,000
you want to amplify your big
input signal how do you do it?

305
00:22:58,000 --> 00:23:02,000
So the question is let's say I
have an input that is this big

306
00:23:02,000 --> 00:23:06,000
here, if it is this big,
I have shown you how I can get

307
00:23:06,000 --> 00:23:11,000
things that are this big,
but what if my input was this

308
00:23:11,000 --> 00:23:13,000
big?
How do I get an output that is

309
00:23:13,000 --> 00:23:16,000
this big?
Well, I will use one of those

310
00:23:16,000 --> 00:23:21,000
learned by questioning methods
and have you tell me the answer.

311
00:23:21,000 --> 00:23:24,000
Someone tell me the answer.
How do I do that?

312
00:23:24,000 --> 00:23:28,000
Yes.
Use another amplifier.

313
00:23:28,000 --> 00:23:34,000
So the answer is I will use one
amplifier to go from here to

314
00:23:34,000 --> 00:23:37,000
here.
And the suggestion is use

315
00:23:37,000 --> 00:23:41,000
another amplifier to go from
here to here.

316
00:23:41,000 --> 00:23:45,000
And, in fact,
I believe that you may have a

317
00:23:45,000 --> 00:23:50,000
problem in your problem set
where you will do that.

318
00:23:50,000 --> 00:23:54,000
And so you have only yourselves
to blame.

319
00:23:54,000 --> 00:24:01,000
So how do you make this work?
What you have to do is this VI

320
00:24:01,000 --> 00:24:05,000
has to be much smaller than the
bias point VI on this one.

321
00:24:05,000 --> 00:24:09,000
I have to build a different
amplifier, choose a different

322
00:24:09,000 --> 00:24:14,000
set of parameters such that VI
prime, which is the VI for this

323
00:24:14,000 --> 00:24:18,000
guy, is much less than V capital
I prime for this guy.

324
00:24:18,000 --> 00:24:22,000
It's a design question.
You need to design it in a way

325
00:24:22,000 --> 00:24:26,000
that the signals of interest
need to be much smaller than the

326
00:24:26,000 --> 00:24:32,000
bias voltage of this amplifier.
So you may have to use much

327
00:24:32,000 --> 00:24:34,000
higher supply voltages.
My amplifier,

328
00:24:34,000 --> 00:24:38,000
I believe, has a 4 volt supply
or 5 volt supply.

329
00:24:38,000 --> 00:24:42,000
You might have to use an
amplifier with a much bigger

330
00:24:42,000 --> 00:24:45,000
supply, different values of RL
and so on.

331
00:24:45,000 --> 00:24:49,000
And I know that the course
notes also have some exercises

332
00:24:49,000 --> 00:24:53,000
and problem sets that discuss
that in more detail.

333
00:24:53,000 --> 00:24:55,000
Yes.
This is even more fun.

334
00:24:55,000 --> 00:24:59,000
The question is,
good question.

335
00:24:59,000 --> 00:25:03,000
The question is why do you need
this guy here?

336
00:25:03,000 --> 00:25:05,000
Just use this guy,
right?

337
00:25:05,000 --> 00:25:09,000
Why do you need this guy?
Big guys rule,

338
00:25:09,000 --> 00:25:13,000
right?
Who needs the little guys?

339
00:25:13,000 --> 00:25:17,000
Well, let me use the Socratic
method again.

340
00:25:17,000 --> 00:25:20,000
Why don't you give me the
answer?

341
00:25:20,000 --> 00:25:25,000
You guys are smart.
Why do you need little guys?

342
00:25:25,000 --> 00:25:30,000
Why do you need the small guy
here?

343
00:25:30,000 --> 00:25:34,000
Anybody with the answer?
Yeah.

344
00:25:34,000 --> 00:25:39,000
The big guy may not be as
sensitive.

345
00:25:39,000 --> 00:25:43,000
I like that.
You know what?

346
00:25:43,000 --> 00:25:50,000
He is almost correct.
I will show you why in a

347
00:25:50,000 --> 00:25:54,000
second.
Anything else?

348
00:25:54,000 --> 00:25:59,000
Any other reason?
Yes.

349
00:26:08,000 --> 00:26:10,000
Bingo.
That is another good answer.

350
00:26:10,000 --> 00:26:13,000
So let me address both the
answers.

351
00:26:13,000 --> 00:26:18,000
The answer given was that look,
this amplifier is amplifying

352
00:26:18,000 --> 00:26:22,000
the signal by a certain amount,
by a factor of 7.

353
00:26:22,000 --> 00:26:27,000
And I have designed this such
that this amplifies a signal by

354
00:26:27,000 --> 00:26:31,000
a factor of maybe 10.
So in all I am getting an

355
00:26:31,000 --> 00:26:34,000
amplification of 70.
This would be a great design

356
00:26:34,000 --> 00:26:37,000
question for lab next year.
I give you a bunch of

357
00:26:37,000 --> 00:26:40,000
components and ask you to design
an amplifier given the

358
00:26:40,000 --> 00:26:43,000
constraints with the highest
amount of amplification.

359
00:26:43,000 --> 00:26:46,000
It turns out that when you
design your amplifier,

360
00:26:46,000 --> 00:26:50,000
in order to meet the saturation
discipline and so on,

361
00:26:50,000 --> 00:26:53,000
you have to choose values of RL
and VS and stuff like that and

362
00:26:53,000 --> 00:26:57,000
be within power constraints so
the amplifier doesn't blow up

363
00:26:57,000 --> 00:27:00,000
and stuff.
And by the end of it all you

364
00:27:00,000 --> 00:27:02,000
are going to get a measly 7X
gain out of it.

365
00:27:02,000 --> 00:27:05,000
The same way here,
to be able to deal with a very

366
00:27:05,000 --> 00:27:08,000
small signal here and get some
amplification,

367
00:27:08,000 --> 00:27:10,000
another set of values and you
get 10X.

368
00:27:10,000 --> 00:27:12,000
So they multiply.
It is much harder to build one

369
00:27:12,000 --> 00:27:14,000
amplifier with a much larger
gain.

370
00:27:14,000 --> 00:27:17,000
You know what?
I just realized that we will be

371
00:27:17,000 --> 00:27:20,000
looking at this in the last five
or seven minutes of lecture.

372
00:27:20,000 --> 00:27:23,000
I am going to show you what the
amplification depends upon.

373
00:27:23,000 --> 00:27:25,000
It depends upon K.
It depends upon RL.

374
00:27:25,000 --> 00:27:30,000
It depends upon VI.
Now the question is I have had

375
00:27:30,000 --> 00:27:33,000
all this time to think about how
to stitch in sensitive into

376
00:27:33,000 --> 00:27:37,000
this, and I believe I can.
It turns out that when you have

377
00:27:37,000 --> 00:27:41,000
large voltages and so on and you
have practical devices,

378
00:27:41,000 --> 00:27:45,000
it turns out that the more
current you pump through devices

379
00:27:45,000 --> 00:27:48,000
they tend to produce noise of
various kinds.

380
00:27:48,000 --> 00:27:52,000
So very powerful amplifiers are
not very good at dealing with

381
00:27:52,000 --> 00:27:55,000
really tiny signals because they
have some inherent noise

382
00:27:55,000 --> 00:27:58,000
capabilities.
And so I guess that is

383
00:27:58,000 --> 00:28:03,000
sensitive.
It is sensitive to noise.

384
00:28:03,000 --> 00:28:07,000
Another question?
Yes.

385
00:28:22,000 --> 00:28:24,000
Ask me the question again.
I didn't follow.

386
00:28:35,000 --> 00:28:38,000
Let me just explain it.
It turns out that I will not be

387
00:28:38,000 --> 00:28:41,000
able to pass this through the
big amplifier to begin with

388
00:28:41,000 --> 00:28:45,000
because it is just going to give
me a gain of just a factor of 7.

389
00:28:45,000 --> 00:28:49,000
However, if I have a signal
that is this big to begin with

390
00:28:49,000 --> 00:28:51,000
then I may just need this
amplifier.

391
00:28:51,000 --> 00:28:54,000
I don't need the smaller guy.
If my signal was this big to

392
00:28:54,000 --> 00:28:58,000
begin with, if I had a strong
sensor that produced a strong

393
00:28:58,000 --> 00:29:01,000
signal to begin with,
yeah, I can deal with just a

394
00:29:01,000 --> 00:29:04,000
single stage.
I don't need to two stages.

395
00:29:04,000 --> 00:29:09,000
It is all a matter of design.
And it is actually a fun design

396
00:29:09,000 --> 00:29:10,000
exercise.
Given a budget,

397
00:29:10,000 --> 00:29:13,000
dollars, right?
You go to your supply room and

398
00:29:13,000 --> 00:29:18,000
look at the parts that you have
and you go to build what you

399
00:29:18,000 --> 00:29:20,000
have to build with the parts
that you have.

400
00:29:20,000 --> 00:29:25,000
And so sometimes you need to
build two amplifiers to get the

401
00:29:25,000 --> 00:29:27,000
gain or build a signal
amplifier.

402
00:29:27,000 --> 00:29:30,000
It's all a design thing.
All right.

403
00:29:30,000 --> 00:29:34,000
Moving on to Page 7.
That brings us to the small

404
00:29:34,000 --> 00:29:36,000
signal model.

405
00:29:51,000 --> 00:29:58,000
Page 5.
What I showed you up on the

406
00:29:58,000 --> 00:30:05,000
little demo was that provided
the signal input in this example

407
00:30:05,000 --> 00:30:11,000
VI was much smaller than capital
VI out there as I shrank my

408
00:30:11,000 --> 00:30:17,000
input, I was able to get a more
or less linear response at the

409
00:30:17,000 --> 00:30:21,000
output.
And so to repeat my notation at

410
00:30:21,000 --> 00:30:28,000
the input, the total input is a
sum of the operating point input

411
00:30:28,000 --> 00:30:35,000
plus a small signal input.
This is called the total

412
00:30:35,000 --> 00:30:41,000
variable.
This is called the DC bias.

413
00:30:41,000 --> 00:30:47,000
It is also called the operating
point voltage.

414
00:30:47,000 --> 00:30:53,000
And this is called my small
signal input.

415
00:30:53,000 --> 00:31:02,000
It is also variously called
incremental input.

416
00:31:02,000 --> 00:31:06,000
This is more a mathematical
term relating to incremental

417
00:31:06,000 --> 00:31:09,000
analysis or perturbation
analysis.

418
00:31:09,000 --> 00:31:14,000
So VI, call it small signal,
call it small perturbation,

419
00:31:14,000 --> 00:31:17,000
call it increment,
whatever you want.

420
00:31:17,000 --> 00:31:23,000
Similarly, at the output I have
my total variable at the output

421
00:31:23,000 --> 00:31:28,000
a sum of the output operating
voltage and the small signal

422
00:31:28,000 --> 00:31:32,000
voltage.
I do not like using Os in

423
00:31:32,000 --> 00:31:39,000
symbols because big O and small
O is simply a function of how

424
00:31:39,000 --> 00:31:43,000
big you write them.
It is not super clear.

425
00:31:43,000 --> 00:31:48,000
And in terms of a graph,
let me plot the input and

426
00:31:48,000 --> 00:31:53,000
output for you.
Let's say this is the total

427
00:31:53,000 --> 00:31:57,000
input and that is the total
output.

428
00:31:57,000 --> 00:32:03,000
I may have some bias VI.
And corresponding to that I may

429
00:32:03,000 --> 00:32:07,000
have some bias VO.
Hold that thought for a second

430
00:32:07,000 --> 00:32:12,000
while I give you a preview of
something that we will be

431
00:32:12,000 --> 00:32:15,000
covering in about three or four
weeks.

432
00:32:15,000 --> 00:32:19,000
Notice that as I couple
amplifiers together,

433
00:32:19,000 --> 00:32:23,000
the output operating point
voltage of this amplifier in

434
00:32:23,000 --> 00:32:28,000
this connection becomes the
input operating point voltage of

435
00:32:28,000 --> 00:32:32,000
this amplifier,
right?

436
00:32:32,000 --> 00:32:34,000
So when they connect this
output to this input,

437
00:32:34,000 --> 00:32:38,000
the output operating point
voltage becomes coupled to the

438
00:32:38,000 --> 00:32:42,000
input here so it becomes the
input operating point voltage

439
00:32:42,000 --> 00:32:44,000
here.
Now I have a nightmare on my

440
00:32:44,000 --> 00:32:46,000
hands.
As I adjust the bias of this

441
00:32:46,000 --> 00:32:48,000
guy, the bias of this guy
changes, too.

442
00:32:48,000 --> 00:32:51,000
The two are dependent.
It is a pain in the neck.

443
00:32:51,000 --> 00:32:55,000
And we being engineers find
ways to simplify our lives.

444
00:32:55,000 --> 00:32:58,000
And you will learn another
trick in about three or four

445
00:32:58,000 --> 00:33:02,000
weeks.
And that trick will let you

446
00:33:02,000 --> 00:33:07,000
decouple these two stages in a
way that you can design this

447
00:33:07,000 --> 00:33:11,000
stage in isolation,
go have a cup of coffee and

448
00:33:11,000 --> 00:33:16,000
then come back to this stage and
design this stage in isolation.

449
00:33:16,000 --> 00:33:22,000
For those of you who want to
run ahead and think about how to

450
00:33:22,000 --> 00:33:26,000
do it, think about it.
What trick can you use to get

451
00:33:26,000 --> 00:33:30,000
them in isolation?
Moving on.

452
00:33:30,000 --> 00:33:35,000
What I would like to do next is
address this from a mathematical

453
00:33:35,000 --> 00:33:39,000
point of view.
And much as I did for the light

454
00:33:39,000 --> 00:33:44,000
emitting expo dweeb analyze this
mathematically and show you that

455
00:33:44,000 --> 00:33:48,000
if VI is much smaller than
capital VI, I indeed get a

456
00:33:48,000 --> 00:33:52,000
linear response.
This time around I won't use

457
00:33:52,000 --> 00:33:57,000
Taylor series because it turns
out that this expression can be

458
00:33:57,000 --> 00:34:02,000
expanded fully.
So you don't have to buy into

459
00:34:02,000 --> 00:34:07,000
Taylor series and so on.
I am going to list everything

460
00:34:07,000 --> 00:34:11,000
down for you.
We know, to begin with,

461
00:34:11,000 --> 00:34:15,000
that VO for the amplifier is
VS-RLK/2 (VI-VT)^2.

462
00:34:15,000 --> 00:34:20,000
What I am going to do for this,
much as I did for the LED,

463
00:34:20,000 --> 00:34:26,000
what I'm going to do is derive
for you the output as a function

464
00:34:26,000 --> 00:34:32,000
of the input when the input VI
is very small.

465
00:34:32,000 --> 00:34:36,000
In other words,
when I substitute for VI,

466
00:34:36,000 --> 00:34:39,000
V capital I squared plus small
VI.

467
00:34:39,000 --> 00:34:46,000
Much as I did for the expo
dweeb, I want to substitute for

468
00:34:46,000 --> 00:34:50,000
VI a big DC VI.
So VI is much smaller than VI.

469
00:34:50,000 --> 00:34:56,000
And show you for yourselves
that the output response,

470
00:34:56,000 --> 00:35:03,000
V small O is going to be
linearly connected to VI.

471
00:35:03,000 --> 00:35:06,000
Notice that,
let me write another equation

472
00:35:06,000 --> 00:35:09,000
here.
This is a total variable.

473
00:35:09,000 --> 00:35:14,000
This simply says that if the
input is VI then the output is

474
00:35:14,000 --> 00:35:18,000
going to be VO,
which means that the operating

475
00:35:18,000 --> 00:35:22,000
point input voltage should
satisfy this equation,

476
00:35:22,000 --> 00:35:24,000
correct?
In other words,

477
00:35:24,000 --> 00:35:30,000
the operating point output
voltage V capital O should equal

478
00:35:30,000 --> 00:35:35,000
VS-RLK/2 (VI-VT)^2.
This is at VI equals capital

479
00:35:35,000 --> 00:35:38,000
VI.
This is very simple but may

480
00:35:38,000 --> 00:35:42,000
seem confusing.
All this is saying is that

481
00:35:42,000 --> 00:35:49,000
look, this equation gives me the
relationship between VI and VO.

482
00:35:49,000 --> 00:35:53,000
Therefore, if I apply capital
VI as the input,

483
00:35:53,000 --> 00:35:58,000
I'm given that my corresponding
output is capital VO,

484
00:35:58,000 --> 00:36:04,000
so they must satisfy this
equation, right?

485
00:36:04,000 --> 00:36:10,000
Those are bias point values and
that must satisfy this equation.

486
00:36:10,000 --> 00:36:12,000
Simple.
I know that.

487
00:36:12,000 --> 00:36:18,000
So hold that thought.
Stash it away in the back of

488
00:36:18,000 --> 00:36:22,000
your minds.
Now let me go through a bunch

489
00:36:22,000 --> 00:36:30,000
of grubby math and substitute
for VI in this expression here.

490
00:36:30,000 --> 00:36:35,000
Let me go ahead and do that.
VS-RLK/2((VI+vi)-VT)^2.

491
00:36:35,000 --> 00:36:41,000
When I do something that is
other than math I will wake you

492
00:36:41,000 --> 00:36:45,000
up.
I will just keep doing a bunch

493
00:36:45,000 --> 00:36:49,000
of steps that are pure math.
No cheating.

494
00:36:49,000 --> 00:36:52,000
No nothing.
Watch my fingers.

495
00:36:52,000 --> 00:37:00,000
When I do anything that is not
obvious math I will wake you up.

496
00:37:00,000 --> 00:37:06,000
Next I am going to simply move
VT over and rewrite this as

497
00:37:06,000 --> 00:37:13,000
follows, RLK/2((VI-VT)+vi)^2.
Again, I haven't done anything

498
00:37:13,000 --> 00:37:18,000
interesting so far.
I have just substituted this.

499
00:37:18,000 --> 00:37:25,000
I am just juggling things
around just to pass away some

500
00:37:25,000 --> 00:37:29,000
time, I guess.
All right.

501
00:37:29,000 --> 00:37:42,000
Next what I am going to do is
simply expand this out and write

502
00:37:42,000 --> 00:37:52,000
it this way RLK/2,
expand that out and treat this

503
00:37:52,000 --> 00:37:59,000
as one unit VS -
RLK/2((VI-VT)^2+

504
00:37:59,000 --> 00:38:06,000
2(VI-VT)vi+vi^2).
Nothing fancy here.

505
00:38:06,000 --> 00:38:11,000
This is like the honest board.
Nothing fancy here.

506
00:38:11,000 --> 00:38:14,000
Standard stuff.
Only math.

507
00:38:14,000 --> 00:38:20,000
I will move to this blackboard
here where I do some fun EE

508
00:38:20,000 --> 00:38:23,000
stuff.
Yes.

509
00:38:28,000 --> 00:38:32,000
Good.
At least one person isn't

510
00:38:32,000 --> 00:38:34,000
asleep here.
Thank you.

511
00:38:34,000 --> 00:38:38,000
So just math here.
Nothing fancy.

512
00:38:38,000 --> 00:38:42,000
Plain old simple math.
I have not done any trickery.

513
00:38:42,000 --> 00:38:45,000
I still have all my ten
fingers.

514
00:38:45,000 --> 00:38:49,000
Now what I am going to do,
now watch me.

515
00:38:49,000 --> 00:38:55,000
I am not using Taylor series
here because this expression

516
00:38:55,000 --> 00:39:00,000
lends itself to this analysis.
Notice VI squared here.

517
00:39:00,000 --> 00:39:06,000
I made the assumption that VI
is much smaller than capital VI,

518
00:39:06,000 --> 00:39:11,000
so what I can do is assuming
that VT is small enough that VI

519
00:39:11,000 --> 00:39:15,000
minus VT is still a big number
compared to small VI,

520
00:39:15,000 --> 00:39:20,000
what I can do is ignore this in
comparison to the capital VI

521
00:39:20,000 --> 00:39:23,000
terms.
So I have a capital VI term

522
00:39:23,000 --> 00:39:26,000
here.
I am going to ignore VI

523
00:39:26,000 --> 00:39:29,000
squared.
So, for example,

524
00:39:29,000 --> 00:39:35,000
if capital VI was 5 volts and
small VI was 100 millivolts 0.1,

525
00:39:35,000 --> 00:39:40,000
so 0.1 squared is 0.01.
So it is comparing 0.01 to 5.

526
00:39:40,000 --> 00:39:44,000
So I am off by a factor of 500.
So now watch me.

527
00:39:44,000 --> 00:39:48,000
Now I begin playing some fun
and games here.

528
00:39:48,000 --> 00:39:52,000
I eliminate this,
and because I eliminate that it

529
00:39:52,000 --> 00:40:02,000
now becomes approximately equal.
What I do in addition is let me

530
00:40:02,000 --> 00:40:10,000
write down the output.
The total variable is the sum

531
00:40:10,000 --> 00:40:18,000
of the DC bias and some
variation of the output.

532
00:40:18,000 --> 00:40:27,000
And let me simply expand that
term and write it down again.

533
00:40:27,000 --> 00:40:34,000
VS-RLK/2(VI-VT)^2-RLK/2.
I get a two here.

534
00:40:34,000 --> 00:40:38,000
And I get VI-VT.
I won't forget the VI this

535
00:40:38,000 --> 00:40:41,000
time.
Again, from here to there

536
00:40:41,000 --> 00:40:45,000
nothing fancy.
This is the one step where I

537
00:40:45,000 --> 00:40:49,000
have used a trick.
I have said small VI is much

538
00:40:49,000 --> 00:40:54,000
smaller than capital VI,
and so I have simply expanded

539
00:40:54,000 --> 00:40:59,000
this out and written it here.
So do you see the obvious next

540
00:40:59,000 --> 00:41:07,000
trick here?
From star look at this guy.

541
00:41:15,000 --> 00:41:20,000
I can cancel this out from star
because I know that at the

542
00:41:20,000 --> 00:41:25,000
operating point these two
expressions are equal,

543
00:41:25,000 --> 00:41:31,000
and so therefore I can cancel
out the operating point voltage

544
00:41:31,000 --> 00:41:38,000
and this.
What I am left with is small VO

545
00:41:38,000 --> 00:41:45,000
is simply minus RLK(VI-VT) times
vi.

546
00:41:45,000 --> 00:41:52,000
Only one place where I did
something funny.

547
00:41:52,000 --> 00:42:00,000
Other than that it is purely
math.

548
00:42:00,000 --> 00:42:05,000
So this is what I get.
Notice that this whole thing is

549
00:42:05,000 --> 00:42:11,000
a constant, minus RLK(VI-VT).
This whole thing is a constant.

550
00:42:11,000 --> 00:42:15,000
And so VO is equal to some
constant times VI.

551
00:42:15,000 --> 00:42:21,000
Let me just define some terms
for you that you will use again

552
00:42:21,000 --> 00:42:25,000
and again.
For reasons that will be

553
00:42:25,000 --> 00:42:30,000
obvious next lecture,
I am going to call this term

554
00:42:30,000 --> 00:42:32,000
here GM.

555
00:42:37,000 --> 00:42:42,000
I am going to call this term a
constant, K(VI - VT).

556
00:42:42,000 --> 00:42:47,000
It is a constant for a given
bias point voltage.

557
00:42:47,000 --> 00:42:53,000
So I am going to call that GM.
And then I am going to call

558
00:42:53,000 --> 00:42:58,000
this whole thing A.
And of course this is VI.

559
00:42:58,000 --> 00:43:03,000
There you go.
I have my linear amplifier.

560
00:43:03,000 --> 00:43:09,000
A is the gain times small VI.
And the gain has these terms in

561
00:43:09,000 --> 00:43:12,000
it.
I just call this GM.

562
00:43:12,000 --> 00:43:17,000
You will see why later.
But notice that the gain

563
00:43:17,000 --> 00:43:21,000
relates to RL.
The size of the load resistor

564
00:43:21,000 --> 00:43:25,000
RL, how big it is,
1K, 10K, whatever.

565
00:43:25,000 --> 00:43:31,000
K, this is a MOSFET parameter,
and VI minus VT.

566
00:43:31,000 --> 00:43:36,000
That is a constant for a given
bias point voltage and small VI.

567
00:43:36,000 --> 00:43:39,000
So VO equals small VI.

568
00:43:47,000 --> 00:43:50,000
I won't give you a graphical
interpretation,

569
00:43:50,000 --> 00:43:55,000
but I encourage you to go and
look at Figure 8.9 in the course

570
00:43:55,000 --> 00:43:57,000
notes.
And it gives you a graphical

571
00:43:57,000 --> 00:44:01,000
interpretation of that
expression.

572
00:44:01,000 --> 00:44:06,000
Move to Page 7.
Another way of looking at this,

573
00:44:06,000 --> 00:44:10,000
another way of mathematically
analyzing it,

574
00:44:10,000 --> 00:44:17,000
here I went through a full
blown expansion and pretty much

575
00:44:17,000 --> 00:44:20,000
deriving the small signal
response.

576
00:44:20,000 --> 00:44:25,000
What I can also do is take a
shortcut here.

577
00:44:25,000 --> 00:44:29,000
So let me just give you the
shortcut.

578
00:44:29,000 --> 00:44:36,000
You might find this handy.
VO=VS-KRL/2(VI-VT)^2.

579
00:44:36,000 --> 00:44:41,000
And my shortcut is as follows.
My small signal response is

580
00:44:41,000 --> 00:44:46,000
simply this relationship.
I find the slope at the point

581
00:44:46,000 --> 00:44:50,000
capital VI and multiply by the
increment.

582
00:44:50,000 --> 00:44:56,000
Slope times the increment gives
me the incremental change in VO

583
00:44:56,000 --> 00:45:01,000
as follows.
d/dI (VS-KRL/2(VI-VT)^2)

584
00:45:01,000 --> 00:45:07,000
evaluated at vI=VI times vi.
This is math again.

585
00:45:07,000 --> 00:45:14,000
I want to find out the change
in VO for a small change in VI,

586
00:45:14,000 --> 00:45:21,000
and I do that by taking the
first derivative of this with

587
00:45:21,000 --> 00:45:29,000
respect to VI substituting V
capital I and multiplying by the

588
00:45:29,000 --> 00:45:35,000
small change delta VI or small
VI.

589
00:45:35,000 --> 00:45:41,000
So this is simply the slope of
the VO versus VI curve at VI.

590
00:45:41,000 --> 00:45:47,000
And so therefore taking the
derivative here of this.

591
00:45:47,000 --> 00:45:51,000
This is a constant so it
vanishes.

592
00:45:51,000 --> 00:45:57,000
But twice 2 to cancel out,
so I get KRL(VI-VT) times small

593
00:45:57,000 --> 00:46:06,000
vi evaluated at capital VI.
So I get twice KRL,

594
00:46:06,000 --> 00:46:18,000
VI evaluated at capital VI,
so it is VI minus VT times

595
00:46:18,000 --> 00:46:23,000
small VI.
Same thing.

596
00:46:23,000 --> 00:46:32,000
Oh, and I have a minus sign
here.

597
00:46:32,000 --> 00:46:37,000
I get the same expression that
I derived for you up there,

598
00:46:37,000 --> 00:46:42,000
and this is just taking the
slope and going with it.

599
00:46:42,000 --> 00:46:46,000
And this, as I mentioned
before, this is A.

600
00:46:46,000 --> 00:46:52,000
The last few minutes let me
kind of pull everything together

601
00:46:52,000 --> 00:46:57,000
and also hit upon something that
many of your questions are

602
00:46:57,000 --> 00:47:02,000
touched upon.
And that all relates to how to

603
00:47:02,000 --> 00:47:07,000
choose the bias point.
So here I have taken an

604
00:47:07,000 --> 00:47:11,000
analysis approach.
When teaching we often teach

605
00:47:11,000 --> 00:47:15,000
you are given something,
you analyze it,

606
00:47:15,000 --> 00:47:20,000
but as you begin to master it
you can begin to design things

607
00:47:20,000 --> 00:47:25,000
where you can ask a lot of
questions and so on.

608
00:47:25,000 --> 00:47:30,000
And here what we have is an
analysis given a value of RLK,

609
00:47:30,000 --> 00:47:34,000
VI and so on.
How to choose the bias point

610
00:47:34,000 --> 00:47:38,000
becomes more of a design issue.
If you are designing an

611
00:47:38,000 --> 00:47:42,000
amplifier, you asked me the
question, how do I choose two

612
00:47:42,000 --> 00:47:46,000
small amplifiers versus one big
amplifier, that sort of stuff?

613
00:47:46,000 --> 00:47:50,000
It boils down to how do you
choose the bias point?

614
00:47:50,000 --> 00:47:54,000
How do you choose VI?
How do you choose RL and so on?

615
00:47:54,000 --> 00:47:59,000
What I would like to do is
touch upon some of these things.

616
00:47:59,000 --> 00:48:02,000
First of all,
gain or the amplification.

617
00:48:02,000 --> 00:48:06,000
One of the most important
design perimeters for an

618
00:48:06,000 --> 00:48:11,000
amplifier is what is the gain?
Let's say you get a job at

619
00:48:11,000 --> 00:48:16,000
Maxim Integrated Technologies,
and they say we would like you

620
00:48:16,000 --> 00:48:20,000
to build a linear power
amplifier for cell phones.

621
00:48:20,000 --> 00:48:23,000
You can say I know how to do
that.

622
00:48:23,000 --> 00:48:28,000
And then they say the next
stage needs a 100 millivolt

623
00:48:28,000 --> 00:48:32,000
input.
While this thing coming from

624
00:48:32,000 --> 00:48:36,000
the antenna is only a few tens
or a few hundreds of a

625
00:48:36,000 --> 00:48:39,000
microvolt.
So you sit down and say oh,

626
00:48:39,000 --> 00:48:43,000
my gosh, I need an
amplification of so much,

627
00:48:43,000 --> 00:48:48,000
and you go design an amplifier.
So gain tends to be a key

628
00:48:48,000 --> 00:48:51,000
parameter.
And notice that gain is

629
00:48:51,000 --> 00:48:55,000
proportional to RL.
It relates to VI minus VT,

630
00:48:55,000 --> 00:49:00,000
so proportional to VI.
It is also related to RL.

631
00:49:00,000 --> 00:49:07,000
The second point is the gain
point determines where I bias

632
00:49:07,000 --> 00:49:12,000
something.
If I choose my bias too high I

633
00:49:12,000 --> 00:49:17,000
get distortion,
or if I choose my bias too low

634
00:49:17,000 --> 00:49:20,000
I get distortion.

635
00:49:27,000 --> 00:49:30,000
So depending on how I choose my
bias point, as a signal goes up

636
00:49:30,000 --> 00:49:33,000
it may begin clipping or begin
distorting.

637
00:49:33,000 --> 00:49:38,000
And I will show you a demo the
next time on that particular

638
00:49:38,000 --> 00:49:41,000
example.
So bias point will determine

639
00:49:41,000 --> 00:49:47,000
how big of a signal you can send
without getting too much

640
00:49:47,000 --> 00:49:50,000
distortion.
And the other thing is that,

641
00:49:50,000 --> 00:49:56,000
relates to how big of an input,
what is a valid input range?

642
00:49:56,000 --> 00:50:02,000
So let's say you have a signal.
And you want that signal to

643
00:50:02,000 --> 00:50:08,000
have both positive and negative
excursions of the same value.

644
00:50:08,000 --> 00:50:12,000
Then, depending on where you
choose a bias point,

645
00:50:12,000 --> 00:50:16,000
your input range may become
smaller or larger.

646
00:50:16,000 --> 00:50:22,000
And we will go through these in
the context of and amplifier and

647
00:50:22,000 --> 00:50:25,000
look at some design issues in
the next lecture.