3/16/2011 section 5_6 Small Signal Operation and Models 1/2

5.6 Small-Signal Operation and Models

Reading Assignment: 443-458

Now let’s examine how we use BJTs to construct amplifiers!

The first important design rule is that the BJT must be biased to
the active mode.

HO: BJT GAIN AND THE ACTIVE REGION

For a BJT amplifier, we find that every current and every voltage
has two components: the DC (i.e., bias) component—a value
carefully selected and designed by a EE, and the small-signal
component, which is the AC signal we are attempting to amplify
(e.g., audio, video, etc.).

HO:DC AND SMALL-SIGNAL COMPONENTS

There are two extremely important circuit elements in small-signal

amplifier design: the Capacitor of Unusual Size (COUS) and the
Inductor of Unusual Size (IOUS).

These devices are just realizable approximations of the

Unfathomably Large Capacitor (ULC) and the Unfathomably
Large Inductor (ULI). These devices have radically different
properties when considering DC and small-signal components!

HO:DC AND AC IMPEDANCE OF REACTIVE ELEMENTS

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 section 5_6 Small Signal Operation and Models 2/2

It turns out that separating BJT currents and voltages into DC and
small-signal components is problematic!

HO:THE SMALL-SIGNAL CIRCUIT EQUATIONS

But, we can approximately determine the small-signal components

if we use the small-signal approximation.

HO: A SMALL-SIGNAL ANALYSIS OF HUMAN GROWTH

HO: A SMALL-SIGNAL ANALYSIS OF A BJT

Let’s do an example to illustrate the small-signal approximation.

EXAMPLE: SMALL-SIGNAL BJT APPROXIMATIONS

There are several small-signal parameters that can be extracted

from a small-signal analysis of a BJT.

HO: BJT SMALL-SIGNAL PARAMETERS

HO: THE SMALL-SIGNAL EQUATION MATRIX

Let’s do an example!

EXAMPLE: CALCULATING THE SMALL-SIGNAL GAIN

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 BJT Gain and the Active Region 1/4

BJT Amplifier Gain and

the Active Region
Consider this simple BJT circuit:
VCC
Q: Oh, goody—you’re
going to waste my time
with another of these
pointless academic RC
problems. Why can’t you
discuss a circuit that
actually does something?
vO

vI
RB

A: Actually, this circuit is a fundamental

electronic device! To see what this circuit
does, plot the output voltage vO as a function of
the input vI. BJT in cutoff
vO
VCC

BJT in active mode

BJT in saturation

vI
VCC
3/16/2011 BJT Gain and the Active Region 2/4

Note that:

vI vO Mode

0 VCC Cutoff

VCC 0 Saturation

Why, this device is not

useless at all! It is clearly a:

____________________

Digital devices made with BJTs typically work in either the

cutoff of saturation regions.

Sir, it appears to
So, what me that the active
good is the region is just a
BJT Active useless BJT mode
Mode ?? between cutoff
and saturation.
3/16/2011 BJT Gain and the Active Region 3/4

Actually, we will find that the active mode is extremely useful!

To see why, take the derivative of the above circuit’s transfer

function (i.e., d VO d VI ):
VO
VCC

d VO
d VI

VI

VCC
We note that in cutoff and saturation:

d VO
≈0
d VI
while in the active mode:

d VO
>> 1
d VI

Q: I’ve got better things to do than

listen to some egghead professor
mumble about derivatives. Are these
results even remotely important?
3/16/2011 BJT Gain and the Active Region 4/4

A: Since in cutoff and saturation d VO d VI = 0 , a small change

in input voltage VI will result in almost no change in output
voltage VO .

Contrast this with the active region, where d VO d VI >> 1 .

This means that a small change in input voltage VI results in a
large change in the output voltageVO !

I see. A small voltage

change results in a big
voltage change—it’s
voltage gain!

The active mode turns

out to be—excellent.

Whereas the important BJT regions for digital devices are

saturation and cutoff, bipolar junction transistors in linear (i.e.,
analog) devices are typically biased to the active region.

This is especially true for BJT amplifier. Almost all of the

transistors in EECS 412 will be in the active region—this is
where we get amplifier gain !
3/16/2011 DC and Small Signal Components 1/5

DC and Small-Signal
Components
Note that we have used DC sources in all of our example circuits
thus far.

We have done this just to simplify the analysis—generally

speaking, realistic (i.e., useful) junction diode circuits will have
sources that are time-varying!

The result will be voltages and currents in the circuit that will
likewise vary with time (e.g., i (t ) and v (t ) ).

For example, we can express the forward bias junction diode

equation as:
vD (t )

iD (t ) = Is e nVT

Although source voltages vS (t ) or currents iS (t ) can be any

general function of time, we will find that often, in realistic and
useful electronic circuits, that the source can be decomposed
into two separate components—the DC component VS , and the
small-signal component v s (t ) . I.E.:

vS (t ) =VS +vs (t )

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and Small Signal Components 2/5

Let’s look at each of these components individually:

* The DC component VS is exactly what you would expect—the

DC component of source vS (t ) !

Note this DC value is not a function of time (otherwise it would

not be DC!) and therefore is expressed as a constant ( e.g.,
VS = 12.3V ).

Mathematically, this DC value is the time-averaged value of

vS (t ) :
T
1
VS = ∫ vS (t ) dt
T 0

where T is the time duration of function vS (t ) .

* As the notation indicates, the small-signal component v s (t ) is

a function of time!

Moreover, we can see that this signal is an AC signal, that is,

its time-averaged value is zero! I.E.:

T
1
T ∫ vs (t )dt = 0
0

This signal v s (t ) is also referred to as the small-signal

component.

* The total signal vS (t ) is the sum of the DC and small signal

components. Therefore, it is neither a DC nor an AC signal!
Jim Stiles The Univ. of Kansas Dept. of EECS
3/16/2011 DC and Small Signal Components 3/5

Pay attention to the notation we have used here. We will use

this notation for the remainder of the course!

* DC values are denoted as upper-case variables

(e.g., VS, IR, or VD).

* Time-varying signals are denoted as lower-case

variables (e.g., vS (t ), v r (t ), iD (t ) ).

Also,

* AC signals (i.e., zero time average) are denoted

with lower-case subscripts (e.g., v s (t ), vd (t ), ir (t ) ).

* Signals that are not AC (i.e., they have a non-

zero DC component!) are denoted with upper-case
subscripts (e.g., vS (t ), ID , iR (t ), VD ).

Note we should never use variables of the form Vi , Ie , Vb . Do

you see why??

Q: You say that we will often find

sources with both components—a DC
and small-signal component. Why is
that? What is the significance or
physical reason for each component?

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and Small Signal Components 4/5

A1: First, the DC component is typically just a DC bias. It is a

known value, selected and determined by the design engineer.

It carries or relates no information—the only reason it exists is

to make the electronic devices work the way we want!

A2: Conversely, the small signal component is typically

unknown!

It is the signal that we are often attempting to process in some

manner (e.g., amplify, filter, integrate). The signal itself
represents information such as audio, video, or data.

Sometimes, however, this small, AC, unknown signal represents

not information—but noise!

Copyright © 2009 Lion Precision. www.lionprecision.com

Noise is a random, unknown signal that in fact masks and

corrupts information.

Our job as designers is to suppress it, or otherwise minimize it

deleterious effects.

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and Small Signal Components 5/5

* This noise may be changing very rapidly with time (e.g.,

MHz), or may be changing very slowly (e.g., mHz).

* Rapidly changing noise is generally “thermal noise”,

whereas slowly varying noise is typically due to slowly
varying environmental conditions, such as temperature.

Note that in addition to (or perhaps because of) the source

voltage vS (t ) having both a DC bias and small-signal component,
all the currents and voltages (e.g., iR (t ), vD (t ) ) within our
circuits will likewise have both a DC bias and small-signal
component!

For example, the junction diode voltage might have the form:

vD (t ) = 0.66 + 0.001cosωt

It is hopefully evident that:

vD VD = 0.66V and vd (t ) = 0.001 cosωt

0.66

t
0.0

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and AC Impedance of Reactive Elements 1/6

DC and AC Impedance
of Reactive Elements
Now, recall from EECS 211 the complex impedances of our
basic circuit elements:

ZR = R

1
ZC =
jωC

Z L = jωL

For a DC signal ( ω = 0 ), we find that:

ZR = R

1
ZC = lim =∞
ω →0 jωC

Z L = j (0)L = 0

Thus, at DC we know that:

* a capacitor acts as an open circuit (IC =0).

* an inductor acts as a short circuit (VL = 0).

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and AC Impedance of Reactive Elements 2/6

Now, let’s consider two important cases:

1. A capacitor whose capacitance C is unfathomably

large.

2. An inductor whose inductance L is unfathomably

large.

1. The Unfathomably Large Capacitor

In this case, we consider a capacitor whose capacitance is

finite, but very, very, very large.

For DC signals (ω = 0 ), this device acts still acts like an open

circuit.

However, now consider the AC signal case (e.g., a small

signal), where ω ≠ 0 . The impedance of an unfathomably
large capacitor is:

1
ZC = lim =0
C →∞ jωC
Zero impedance!

J An unfathomably large capacitor acts like an AC short.

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and AC Impedance of Reactive Elements 3/6

Quite a trick! The unfathomably large capacitance acts like an

open to DC signals, but likewise acts like a short to AC (small)
signals!

+ vc (t ) = 0 −

IC = 0 C = lim C
C →∞

Q: I fail to see the relevance

of this analysis at this juncture.
After all, unfathomably large
capacitors do not exist, and are
impossible to make (being
unfathomable and all).

A: True enough! However, we can make very big (but

fathomably large) capacitors. Big capacitors will not act as a
perfect AC short circuit, but will exhibit an impedance of very
small magnitude (e.g., a few Ohms), provided that the AC
signal frequency is sufficiently large.

In this way, a very large capacitor acts as an approximate AC

short, and as a perfect DC open.

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and AC Impedance of Reactive Elements 4/6

We call these large capacitors DC blocking capacitors, as

they allow no DC current to flow through them, while allowing
AC current to flow nearly unimpeded!

Q: But you just said this is true

“provided that the AC signal
frequency is sufficiently large.”
Just how large does the signal
frequency ω need to be?

A: Say we desire the AC impedance of our capacitor to have a

magnitude of less than ten Ohms:

ZC < 10

Rearranging, we find that this will occur if the frequency ω

is:

10 > ZC
1
10 >
ωC
1
ω >
10C

For example, a 50 µF capacitor will exhibit an impedance

whose magnitude is less than 10 Ohms for all AC signal
frequencies above 320 Hz.

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and AC Impedance of Reactive Elements 5/6

Likewise, almost all AC signals in modern electronics will

operate in a spectrum much higher than 320 Hz.

Thus, a 50 µF blocking capacitor will approximately act as an

AC short and (precisely) act as a DC open.

2. The Unfathomably Large Inductor

Similarly, we can consider an unfathomably large inductor. In

addition to a DC impedance of zero (a DC short), we find for
the AC case (where ω ≠ 0 ):

Z L = lim jωL = ∞
L →∞

In other words, an unfathomably large inductor acts like an

AC open circuit!

+ VC = 0 −

iA (t ) = 0 L = lim L
L →∞

The unfathomably large inductor acts like an short to DC

signals, but likewise acts like an open to AC (small) signals!

Jim Stiles The Univ. of Kansas Dept. of EECS

3/16/2011 DC and AC Impedance of Reactive Elements 6/6

As before, an unfathomably large inductor is impossible to

build.

However, a very large inductor will typically exhibit a very

large AC impedance for all but the lowest of signal
frequencies ω .

We call these large inductors “AC chokes” (also known RF

chokes), as they act as a perfect short to DC signals, yet so
effectively impede AC signals (with sufficiently high
frequency) that they act approximately as an AC open
circuit.

For example, if we desire an AC choke with an impedance

magnitude greater than 100 kΩ, we find that:

Z L > 105
ωL > 105
105
ω >
L

Thus, an AC choke of 50 mH would exhibit an impedance

magnitude of greater than 100 kΩ for all signal frequencies
greater than 320 kHz.

Note that this is still a fairly low signal frequency for many
modern electronic applications, and thus this inductor would
be an adequate AC choke. Note however, that building and AC
choke for audio signals (20 Hz to 20 kHz) is typically very
difficult!

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 The Small Signal Circuit Equations 1/4

The Small-Signal
Circuit Equations
Now let’s again consider this circuit, where we assume the BJT
is in the active mode:
VCC

RC iC

vO
RB
vI
+
iB vBE
−

The four equations describing this circuit are:

1) vI − RB iB − vBE = 0 (KVL)

2) iC = β iB (BJT)

3) vO = VCC − RC iC (KVL)

vBE
4) iC = I s e VT
(BJT)
3/28/2011 The Small Signal Circuit Equations 2/4

Now, we assume that each current and voltage has both a small-
signal and DC component. Writing each equation explicitly in
terms of these components, we find that the four circuit
equations become:

(1) (VI +vi ) − RB (IB + ib ) − (VBE + vbe ) = 0

(VI − RB IB −VBE ) + (vi − RB ib −vbe ) = 0

(2) IC + ic = β (IB + ib )
IC + ic = β IB + β ib

(3) VO +vo =VCC − RC (IC + ic )

VO +vo = (VCC − RC IC ) − RC ic

(VBE +vbe )
(4) IC + ic = Is e VT

VBE vbe
IC + ic = Is e VT
e VT

Note that each equation is really two equations!

1. The sum of the DC components on one side of the equal sign

must equal the sum of the DC components on the other.

2. The sum of the small-signal components on one side of the

equal sign must equal the sum of the small-signal components on
the other.

This result can greatly simplify our quest to determine the

small-signal amplifier parameters!
3/28/2011 The Small Signal Circuit Equations 3/4

You see, all we need to do

is determine four small-
signal equations, and we
can then solve for the
four small-signal values
ib , ic , vbe , vo !

From (1) we find that the DC equation is:

VI − RB IB −VBE = 0

while the small-signal equation from 1) is:

vi − RB ib − vbe = 0

Similarly, from equation (2) we get these equations:

IC = β I B (DC)

ic = β ib (small signal)

And from equation (3):

VO = VCC − RC IC (DC)

vo = RC ic (small-signal)
 3/28/2011 The Small Signal Circuit Equations 4/4

Finally, from equation (4) we, um, get, er—just what the heck do
we get?

(VBE +vbe )
(4) IC + ic = Is e VT

VBE vbe
????
IC + ic = Is e VT
e VT

Q: Jeepers! Just what are the DC

and small-signal components of:

VBE vbe
Is e VT
e VT
???

A: Precisely speaking, we cannot express the above expression

as the sum of a DC and small-signal component. Yet, we must
determine a fourth small-signal equation in order to determine
the four small signal values ib , ic , vbe , vo !

However, we can approximate the above expression as the sum

of DC and small-signal components. To accomplish this, we must
apply the small-signal approximation (essentially a Taylor series
approx.).

We will find that the small-signal approximation provides an

accurate small-signal equation for expressions such (4). We will
likewise find that this approximate equation is accurate if the
small-signal voltage vbe is, well, small!
 3/16/2011 A Small-Signal Analysis of Human Growth 1/6

A “Small-Signal Analysis”
of Human Growth
Say the average height h of a human (in inches) is related to
his/her age t in months by this equation:

h (t ) = 65 − 3. 66x10-10 (45 − t 12 )6.75 inches

h (t )
65 inches

We shrink
when we age!
Those awkward
adolescent years!

t
70 years

Say that we want to calculate the average height of a human at

an age of t =58, 59, 59.5, 60, 60.5, 61, and 62 months.

Whew! Let me get out my calculator!

3/16/2011 A Small-Signal Analysis of Human Growth 2/6

h (t = 58. 0) = 40. 48 inches

h (t = 59. 0) = 40. 82 inches
h (t = 59. 5) = 40. 99 inches
h (t = 60. 0) = 41. 16 inches
h (t = 60. 5) = 41. 32 inches
h (t = 61. 0) = 41. 49 inches
h (t = 62. 0) = 41. 82 inches

Q: Wow, this was hard. Isn’t there an easier way to calculate

these values?

A: Yes, there is! We can make a “small-signal” approximation.

For a small-signal approximation, we simply need to calculate two

values. First:
h (t ) t =60 = h (t = 60) = 41. 16 inches

In other words, the average height of a human at 60 months

(i.e., 5 years) is 41.16 inches.

Likewise, we calculate the time derivative ofh (t ) , and then

evaluate the result at 60 months:

d h (t )
= (2. 059 x 10-10 (45 − t 12 )5.75 )
dt t =60 t = 60

= 2. 059 x 10-10 (45 − 60 12 )5.75

= 0. 34 inches/month
3/16/2011 A Small-Signal Analysis of Human Growth 3/6

In other words, the average 5 year old grows at a rate of 0.34

inches/month!

Now let’s again consider the earlier problem.

If we know that an average 5-year old is 41.16 inches tall, and

grows at a rate of 0.34 inches/month, then at 5 years and one
month (i.e., 61 months), the little bugger will approximately be:

41. 16 + (0.34)(1) = 41. 50 inches

Compare this to the exact value of 41.49 inches—a very

accurate approximation.

We can likewise approximate the average height of a 59-month

old (i.e., 5 years minus one month):

41. 16 + (0.34)( − 1) = 40. 83 inches

or of a 62-month old (i.e., 5 years plus two months):

41. 16 + (0.34)(2) = 41. 83 inches

Note again the accuracy of these approximations!

For this approximation, let us define time t =60 as the

evaluation point, or bias point T :

T  evaluation point
3/16/2011 A Small-Signal Analysis of Human Growth 4/6

We can then define:

∆t = t −T

In this example then, T = 60 months, and the values of ∆t range

from –2 to +2 months.

For example, t = 59 months can be expressed as t = T + ∆t ,

where T = 60 months and ∆t = −1 month.

We can therefore write our approximation as:

d h (t )
h (t ) ≈ h (t ) t =T + ∆t
dt t =T

For the example where T =60 months we find:

d h (t )
h (t ) ≈ h (t ) t =60 + ∆t
dt t =60
= 41. 16 + 0. 34 ∆t

This approximation is not accurate, however, if ∆t is large.

For example, we can determine from the exact equation that

the average height of a forty-year old human is:

h (t = 480) = 65 inches

or about 5 feet 5 inches.

3/16/2011 A Small-Signal Analysis of Human Growth 5/6

However, if we were to use our approximation to determine the

average height of a 40-year old ( ∆t = t −T = 480 − 60 = 420 ),
we would find:

h (t ) ≈ 41. 16 + 0. 34 (420)
= 181. 86 inches

The approximation says that the average 40-year old human is

over 15 feet tall!
Where exactly do I
find these dad-gum
humans?

The reason that the above approximation provides an inaccurate

answer is because it is based on the assumption that humans
grow at a rate of 0.34 inches/month.

This is true for 5-year olds, but not for 40-year

olds (unless, of course, you are referring to
their waistlines)!

We thus refer to the approximation function as a “small-signal”

approximation, as it is valid only for times that are slightly
different from the nominal (evaluation) time T (i.e., ∆t is small).
3/16/2011 A Small-Signal Analysis of Human Growth 6/6

If we wish to have an approximate function for the growth of

humans who are near the age of forty, we would need to
construct a new approximation:

d h (t )
h (t ) ≈ h (t ) t = 480 + ∆t
dt t = 480

= 65. 0 + 2. 2 x 10 −6 ∆t

Note that forty-year old humans have stopped growing!

The mathematically astute will recognize the small-signal model

as a first-order Taylor Series approximation!
3/28/2011 A Small Signal Analysis of a BJT lecture 1/12

A Small-Signal
Analysis of a BJT
The collector current iC of a BJT is related to its base-emitter voltage vBE as:

iC

vBE
iC = IS e VT

vBE

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 2/12

One messy result

Say the current and voltage have both D.C. ( IC , VBE ) and small-signal (ic , vbe )
components:

iC (t ) = IC + ic (t )
and

vBE (t ) = VBE + vbe (t )

Therefore, the total collector current is:

vBE (t )
iC (t ) = IS e VT

VBE +vbe (t )

IC + ic (t ) = IS e VT

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 3/12

Apply the Small-Signal Approximation

Q: Yikes! The exponential term is very messy. Is there some way to

approximate it?

A: Yes! The collector current ic is a function of base emitter voltage vBE.

Let’s perform a small-signal analysis to determine an approximate relationship

between ic and vBE.

Note that the value of vBE (t ) = VBE + vbe (t ) is always very close to the D.C.
voltage for all time t (since vbe (t ) is very small).

We therefore will use this D.C. voltage as the evaluation point (i.e., bias point)
for our small-signal analysis.

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 4/12

How fast it grows!

We first determine the value of the collector current iC when the base emitter
voltage vBE is equal to the DC value VBE :

VBE
vBE
iC vBE =VBE
=I e VT = Is e VT
= IC
s vBE =VBE

Of course, the result is the D.C. collector current IC .

We now determine the change in collector current due to a change in base-

emitter voltage (i.e., a first derivative), evaluated at the D.C. voltage VBE :

d iC d ( IS exp ⎡⎣vBE VT ⎤⎦ )
=
d vBE vBE =VBE
d vBE
vBE =VBE

IS vBE VT
= e
VT vBE =VBE

IS VBE VT ⎡A ⎤
= e ⎣ V⎦
VT

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 5/12

A simple approximation
Thus, when the base-emitter voltage is equal to the D.C. “bias” voltage VBE , the
collector current iC will equal the D.C. “bias” current IC .

Likewise, this collector current will increase (decrease) by an amount of

(IS VT ) eVBE VT mA for every 1mV increase (decrease) in vBE .
vBE = VBE + 1 mV
Thus, we can easily approximate the collector current
vBE = VBE + 3 mV
when the base-emitter voltage is equal to values such as:
vBE = VBE − 2 mV
Respectively, the answers are: vBE = VBE − 0.5 mV

VBE
iC = IC + ( IS VT ) e VT
(1) mA
VBE
iC = IC + ( IS VT ) e VT
(3) mA
VBE
iC = IC + ( IS VT ) e VT
(-2) mA
VBE
iC = IC + ( IS VT ) e VT
(-0.5) mA

where we have assumed that scale current IS is expressed in mA, and thermal
voltage VT is expressed in mV.

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 6/12

The small signal approximation

Recall that the small-signal voltage vbe (t ) represents a small change in vBE (t )
from its nominal (i.e., bias) voltage VBE .

For example, we might find that the value of vbe (t ) at four different times t
are:
vbe (t1 ) = 1 mV
vbe (t2 ) = 3 mV
vbe (t3 ) = −2 mV
vbe (t4 ) = −0.5 mV

Thus, we can approximate the collector current using the small-signal

approximation as:

iC (t ) = IC + ( IS VT ) eVBE VT vbe (t )

where of course IC = IS eVBE VT .

This is a very useful result, as we can now explicitly determine an expression for
the small-signal current ic (t ) !

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 7/12

The small-signal collector current

Recall iC (t ) = IC + ic (t ) , therefore:

iC (t ) = IC + ic (t ) = IC + ( IS VT ) eVBE VT vbe (t )

Subtracting the D.C. current from each side, we are left with an expression for
the small-signal current ic (t ) , in terms of the small-signal voltage vbe (t ) :

ic (t ) = ( IS VT ) eVBE VT vbe (t )

We can simplify this expression by noting that IC = IS eVBE VT , resulting in:

IS eVBE VT
( IS VT ) eVBE VT
=
VT
I
= C
VT
and thus:

IC
ic (t ) = v (t )
VT be

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 8/12

Transconductance: A small signal parameter

We define the value IC VT as the transconductance gm :

IC ⎡A ⎤
gm = ⎣ V⎦
VT

and thus the small-signal equation simply becomes:

ic (t ) = gm vbe (t )

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 9/12

How transistors got their name

Let’s now consider for a moment the transconductance gm .

The term is short for transfer conductance: conductance because its units
are amps/volt, and transfer because it relates the collector current to the
voltage from base to emitter—the collector voltage is not relevant (if in
active mode)!

Note we can rewrite the small-signal equation as:

vbe (t ) 1
=
ic (t ) gm

The value ( 1 gm ) can thus be considered as transfer resistance, the value

describing a transfer resistor.

Transfer Resistor—we can shorten this term to Transistor (this is how these
devices were named)!

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 10/12

Summarizing
We can summarize our results as:

IC = IS eVBE VT D.C. Equation

ic (t ) = gm vbe (t ) Small-Signal Equation

iC (t ) = IC + gm vbe (t ) Small-Signal Approximation

Note that we know have two expressions for the total (D.C. plus small-signal)
collector current. The exact expression:

VBE +vbe (t )

iC (t ) = IS e VT

and the small-signal approximation:

iC (t ) = IC + gm vbe (t )

Jim Stiles The Univ. of Kansas Dept. of EECS

3/28/2011 A Small Signal Analysis of a BJT lecture 11/12

Accurate over a small region

Let’s plot these two expressions and see how they compare:

iC
Exact
Small-signal
Validity Regions

gm

IC

vBE

VBE

It is evident that the small-signal approximation is accurate (it provides nearly

the exact values) only for values of iC near the D.C bias value IC , and only for
values of vC near the D.C bias value VC.

The point (VBE , IC ) is alternately known as the D.C. bias point, the transistor
operating point, or the Q-point.
Jim Stiles The Univ. of Kansas Dept. of EECS
3/28/2011 A Small Signal Analysis of a BJT lecture 12/12

Change the DC bias,

change the transconductance
Note if we change the D.C. bias of a transistor circuit, the transistor
operating point will change.

The small-signal model will likewise change, so that it provides accurate results
in the region of this new operating point:

iC gm
Exact
Small-signal
Validity Regions

IC

vBE

VBE
Jim Stiles The Univ. of Kansas Dept. of EECS
3/30/2011 Example A Small Signal BJT Approximation 1/3

Example: Small-Signal BJT

Approximations
Say that we wish to find the collector current iC of a BJT
biased in the active mode, with IS = 10 −12 A and a base-emitter
voltage of:
vBE = 0.6 + 0.001 cosωt V

Q: Easy! Since:
vBE
iC = IS e VT

we find:

⎛ 0.6
⎞ 0.001cosωt VT )
iC (t ) = ⎜ IS e VT ⎟ e (
⎝ ⎠

right?

A: Although this answer is definitely correct, it is not very

useful to us as engineers. Clearly, the base-emitter voltage
consists of a D.C. bias term (0.6 V) and a small-signal term
( 0.001 cosωt ).

Accordingly, we are interested in the D.C. collector current IC

and the small-signal collector current ic . The D.C. collector
current is obviously:
3/30/2011 Example A Small Signal BJT Approximation 2/3

0.6
IC = I S e VT

0. 6
= 10 −12 e 0.025

= 26 mA

But how do we determine the small-signal collector current

ic (t ) from:
⎛ 0.6
⎞ ( 0.001cosωt VT )
iC (t ) = ⎜ IS e VT ⎟e ???
⎝ ⎠

The answer, of course, is to use the small-signal approximation.

We know that:

ic (t ) = gm vbe (t )
where:

IC 26mA
gm = = = 1.06 Ω −1
VT 25mV

Therefore, the small-signal collector current is approximately:

ic (t ) = gm vbe (t )
= 1.06 ( 0.001 cosωt )
= 1.06cosωt mA

and therefore the total collector current is:

3/30/2011 Example A Small Signal BJT Approximation 3/3

Q: Say the D.C. bias voltage increases from VBE = 0.6 V to

VBE = 0.7 V . What happens to the BJT collector current?

A: The D.C. bias current becomes:

0. 7 0.7
IC = IS e VT
= 10 −12 e 0.025
= 1446 mA !!!

since the transconductance is now:

IC 1446mA
gm = = = 57.84 Ω −1
VT 25mV

the small-signal collector current is:

ic (t ) = gm vbe (t )
= 57.84 ( 0.001 cosωt )
= 57.8cosωt mA

Quite an increase!

Changing the transistor operating point (i.e., the DC bias point)

will typically make a big difference in the small-signal result!
3/30/2011 BJT Small Signal Parameters lecture 1/5

BJT Small-Signal Parameters

We know that the following small-signal relationships are true for BJTs:

ic = β ib ic = gm vbe

Q: What other relationship can be derived from these two??

A: Well, one obvious relationship is determined by equating the two equations

above:

⎛ β ⎞
ic = β ib = gm vbe ∴ vbe = ⎜
⎜g ⎟⎟ ib
⎝ m ⎠

We can thus define the small-signal parameter rπ as:

β βV V
= T = T  rπ
gm IC IB

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 BJT Small Signal Parameters lecture 2/5

Small-signal base resistance

Therefore, we can write the new BJT small-signal equation:

vbe = rπ ib

The value rπ is commonly thought of as the small-signal base resistance.

We can likewise define a small-signal emitter resistance:

vbe
re 
ie

We begin with the small-signal equation ic = α ie . Combining this with ic = gm vbe ,

we find:
⎛ α ⎞
ic = α ie = gm vbe ∴ vbe = ⎜
⎜g ⎟⎟ ie
⎝ m ⎠

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 BJT Small Signal Parameters lecture 3/5

Small-signal emitter resistance

We can thus define the small-signal parameter re as:

α αV V
= T = T  re
gm IC IE

Therefore, we can write another new BJT small-signal equation:

vbe = re ie

Note that in addition to β , we now have three fundamental BJT small-signal

parameters:

IC VT VT
gm = rπ = re =
VT IB IE

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 BJT Small Signal Parameters lecture 4/5

These results are not independent!

Since IC = β IB ( IC = α IE ), we find that these small signal values are not
independent.

If we know two of the four values β , gm , rπ , re , we can determine all four!

α β rπ − re
gm = = =
re rπ rπ re

β re
rπ = = ( β + 1 ) re =
gm 1 − gm re

α rπ rπ
re = = =
gm β + 1 1 + gm rπ

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 BJT Small Signal Parameters lecture 5/5

Make sure you can derive these!

The results on the previous page are easily determined from the equations:

IC VT VT
gm = rπ = re =
VT IB IE

I E = IC + I B

IC = β IB

IC = α IE

Make sure you can derive them!

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 The Small Signal Equaton Matrix lecture 1/2

The Small-Signal Equation Matrix

We can summarize our small-signal equations with the small-signal equation
matrix. Note this matrix relates the small-signal BJT parameters vbe , ib , ic ,
and ie .

Column Parameters

vbe ib ic ie

vbe 1 rπ = β 1
gm re = α g
gm m

Row 1 gm 1 1
ib
Parameters rπ
=
β
1 β (β + 1)
β
ic gm β 1 α=
β +1

1 gm 1 β +1
ie = β +1 = 1
re α α β

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 The Small Signal Equaton Matrix lecture 2/2

Here’s how you use this

To use this matrix, note that the row parameter is equal to the product of the
column parameter and the matrix element. For example:

1
ib = vbe
rπ

vbe ib ic ie

vbe 1 rπ = β 1
gm re = α g
gm m

1 gm 1 1
ib
rπ
=
β
1 β ( β + 1)
β
ic gm β 1 α=
β +1

1 gm 1 β +1
ie = β +1 = 1
re α α β

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 Example Calculating the Small Signal Gain 1/2

Example: Calculating the

Small-Signal Gain
For this circuit, we have now determined (if BJT is in active
mode), the small-signal equations are:
VCC
1) vi = ib RB + vbe

RC iC 2) ic = β ib

3) vo = −Rc ic
vO
RB 4) ic ≅ gm vbe
vI
+
iB vBE
−

Q: So, can we now determine the small-signal open-circuit

voltage gain of this amplifier? I.E.:

vo (t )
Avo =
vi (t )

A: Look at the four small-signal equations—there are four

unknowns (i.e.,ib , vbe , ic , vo )!

Combining equations 2) and 4), we get:

3/30/2011 Example Calculating the Small Signal Gain 2/2

β
vbe = i =ri
gm b π b

Inserting this result in equation 1), we find:

vi = (RB + rπ ) ib

Therefore:
vi
ib =
RB + rπ

and since ic = β ib :
β
ic = vi
RB + rπ

which we insert into equation 3):

− β RC
vo = −ic RC = vi
RB + rπ

Therefore, the small-signal gain of this amplifier is:

vo (t ) − β RC
Avo = =
vi (t ) RB + rπ

Note this is the small signal gain of this amplifier—and this

amplifier only!
3/30/2011 The Hybrid Pi and T Models lecture 1/6

The Hybrid-Π and T Models

Consider again the small-small signal equations for an npn BJT biased in the
active mode:

vbe
ib = ic = gm v be = β ib ie = ib + ic (KCL)
rπ

Now, analyze this circuit:

ib ic

gm vbe
+
= β ib
rπ vbe
-

ie

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 The Hybrid Pi and T Models lecture 2/6

Do these equations look familiar?

From Ohm’s Law:
vbe
ib = ib
rπ ic

gm vbe
From KCL: +
= β ib
ic = gm v be = β ib rπ vbe
-

And also from KCL:

ie = ib + ic (KCL) ie

Q: Hey! Aren’t these the same three equations as the npn BJT small-signal
equations?

A: They are indeed!

With respect to the small-signal currents and voltages in a circuit (but only
small-signal voltages and currents) , an npn BJT in active mode might as well be
this circuit.

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 The Hybrid Pi and T Models lecture 3/6

Two equivalent circuits

Thus, this circuit can be used as an equivalent circuit for BJT small-signal
analysis (but only for small signal analysis!).
ib ic
This equivalent circuit is
gm vbe C
called the Hybrid-Π model B
for a BJT biased in the + = β ib
rπ vbe v be
active mode: ib = rπ
-

npn Hybrid-Π Model ic = gm v be = β ib

ie
E ie = ib + ic

E
v eb
ib = rπ ie

ic = gm v eb = β ib
+ pnp Hybrid-Π Model
rπ veb
ie = ib + ic gm veb
-
ib ic
= β ib
B C

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 The Hybrid Pi and T Models lecture 4/6

An alternative equivalent circuit

Note however, that we can alternatively express the small-signal circuit
equations as:
vbe
ib = ie − ic ic = gm vbe = β ib ie =
re

These equations likewise describes the T-Model—an alternative but equivalent

model to the Hybrid-Π.

C
ic

npn T-Model ib = ie − ic
gm v be
ib = β ib ic = gm v be = β ib
B

+ vbe
ie =
re vbe re
-

ie
E

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 The Hybrid Pi and T Models lecture 5/6

I just couldn’t fit the pnp

T-model on the previous page

E
ie
pnp T-Model
+
ib = ie − ic
re veb
-
ib
ic = gm veb = β ib
B gm v be
= β ib veb
ie =
re

ic
C

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 The Hybrid Pi and T Models lecture 6/6

So many choices; which should I use?

The Hybrid-Π and the T circuit models are equivalent—they both will result in
the same correct answer!

Therefore, you do not need to worry about which one to use

for a particular small-signal circuit analysis, either one will
work.

However, you will find that a particular analysis is easier with one model or the
other; a result that is dependent completely on the type of amplifier being
analyzed.

For time being, use the Hybrid-Π model; later on, we will discuss the types of
amplifiers where the T-model is simplest to use.

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 Small Signal Output Resistance lecture 1/5

Small-Signal Output Resistance

Recall that due to the Early effect, the collector current iC is slightly
dependent on vCE :
⎛ vCE ⎞
iC = β iB ⎜ 1 + ⎟
⎝ VA ⎠

where we recall that VA is a BJT device parameter, called the Early Voltage.

Q: How does this affect the small-signal response of the BJT?

A: Well, if iC (t ) = IC + ic (t ) and vCE (t ) = VCE + vce (t ) , then with the small-signal

approximation:

⎛ vCE ⎞ ⎛ ∂iC ⎞
IC + ic = β iB ⎜ 1 + ⎟ +⎜ ⎟vce
⎝ VA ⎠ ⎜ ∂v ⎟
vCE =VCE ⎝ CE vCE =VCE ⎠
⎛ V ⎞ ⎛ V ⎞⎛ 1 ⎞
= β IB ⎜ 1 + CE ⎟ + β IB ⎜ 1 + CE ⎟ ⎜ ⎟vce
⎝ VA ⎠ ⎝ VA ⎠ ⎝VA ⎠

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 Small Signal Output Resistance lecture 2/5

Small-signal base resistance

Equating the DC components:
⎛ VCE ⎞
IC = β IB ⎜ 1 + ⎟
⎝ VA ⎠

And equating the small-signal components:

⎛ VCE ⎞ ⎛ 1 ⎞
ic = β IB ⎜ 1 + ⎟ ⎜ ⎟v
⎝ VA ⎠ ⎝VA ⎠ ce

Note that by inserting the DC result, this expression can be simplified to:

⎛ 1 ⎞ ⎛I ⎞
ic = IC ⎜ ⎟vce = ⎜ C ⎟vce
⎝VA ⎠ ⎝ VA ⎠

Therefore, another small-signal equation is found, one that expresses the

small-signal response of the Early effect:

⎛ IC ⎞
ic = ⎜ ⎟ vce
V
⎝ A ⎠

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 Small Signal Output Resistance lecture 3/5

Small-signal base resistance

Recall that we defined (in EECS 312) the BJT output resistance ro :

IC 1

VA ro

Be careful! Although the Early Voltage VA is a device parameter,

the output resistance ro —since it depends on DC collector
current IC —is not a device parameter!

Therefore, the small-signal collector current resulting from the Early effect
can likewise be expressed as:

vce
ic =
ro

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 Small Signal Output Resistance lecture 4/5

Small-signal base resistance

Combining this result with an earlier result (i.e., ic = gm vbe ), we find that the
total small-signal collector current is:

vce v
ic = gm vbe + = β ib + ce
ro ro

We can account for this effect in our small-signal circuit models. For example,
the Hybrid-Π becomes:

ib ic v be
ib =
gm v be C rπ
B
+ = β ib +
rπ vbe vce ro ic = gm v be +
vbe
ro
- -

ie = ib + ic

ie
E

Jim Stiles The Univ. of Kansas Dept. of EECS

3/30/2011 Small Signal Output Resistance lecture 5/5

Small-signal base resistance

C
ic

And for the T-model: gm v be

= β ib
ib +
B vce ro
-
+
vbe re
-

ie
E

Often, ro is so large that it can be ignored (caution: ignoring the output

resistance means approximating it as an open circuit, i.e., ro = ∞ ).

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 1/14

BJT Small-Signal
Analysis Steps
Complete each of these steps if you choose to correctly complete a BJT
Amplifier small-signal analysis.

Step 1: Complete a D.C. Analysis

Turn off all small-signal sources, and then complete a circuit analysis with the
remaining D.C. sources only.

* Complete this DC analysis exactly, precisely, the same way you

performed the DC analysis in section 5.4.

That is, you assume (the active mode), enforce, analyze, and check (do not
forget to check!).

* Note that you enforce and check exactly, precisely the same the same
equalities and inequalities as discussed in section 5.4 (e.g., VBE = 0.7 V ,
VCB > 0 ).

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 2/14

You must remember this

* Remember, if we “turn off” a voltage source (e.g.,vi (t ) = 0 ), it becomes a

short circuit.

* However, if we “turn off” a current source (e.g., ii (t ) = 0 ), it becomes an

open circuit!

* Small-signal amplifiers frequently employ

Capacitors of Unusual Sizes (COUS), we’ll discuss
why later.

Remember, the impedance of a capacitor at DC is

infinity—a DC open circuit.

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 3/14

The goals of DC analysis—

and don’t forget to CHECK

The goal of this DC analysis is to determine:

1) One of the DC BJT currents ( IB , IC , IE ) for each BJT.

2) Either the voltage VCB or VCE for each BJT.

You do not necessarily need to determine any other DC currents or

voltages within the amplifier circuit!

Once you have found these values, you can CHECK your active assumption, and
then move on to step 2.

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 4/14

The DC bias terms are required to

determine our small-signal parameters

Q: I’m perplexed. I was eagerly

anticipating the steps for small-
signal analysis, yet you’re saying
we should complete a DC analysis.

Why are we doing this—why do

we care what any of the DC
voltages and currents are?

A: Remember, all of the small-signal BJT parameters (e.g., gm , rπ , re , ro ) are

dependent on D.C. values (e.g., IC , IB , IE ).

In other words, we must first determine the operating (i.e., bias) point of the
transistor in order to determine its small-signal performance!

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 5/14

Now for step 2

Step 2: Calculate the small-signal circuit parameters for each BJT.

Recall that we now understand 4 small-signal parameters:

IC VT VT VA
gm = rπ = re = ro =
VT IB IE IC

Q: Yikes! Do we need to calculate all four?

A: Typically no. You need to calculate only the small signal parameters
required by the small-signal circuit model that you plan to implement.

For example, if you plan to:

a) use the Hybrid-Π model, you must determine gm and rπ .

b) use the T-model, you must determine gm and re .

c) account for the Early effect (in either model) you must determine ro .

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 6/14

The four “Pees”

Step 3: Carefully replace all BJTs with their small-signal circuit model.

This step often gives students fits!

However, it is actually a very simple and straight-forward step.

It does require four important things from the student—patience, precision,

persistence and professionalism!

+ C
vCB + iC
iB
-
First, note that a BJT is: vCE
B
+
vBE -
A device with three terminals, called
-
the base, collector, and emitter.

Its behavior is described in terms of iE

currents iB , iC , iE and voltages E
vBE , vCB , vCE .

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 7/14

They’re both so different—not!

Now, contrast the BJT with its small-signal circuit model.

A BJT small-signal circuit model is:

A device with three terminals, called the base, collector, and emitter.

Its behavior is described in terms of currents ib , ic , ie and voltages

vbe , vcb , vce .

Exactly the same—what a coincidence!

ib ic
B - vcb +
C
+ +

rπ vbe gmvbe vce

-
-

ie
E

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 8/14

Am I making this clear?

Therefore, replacing a BJT with its small-signal circuit model is very simple—you
simply change the stuff within the orange box!

Note the parts of the circuit external to the orange box do not change! In
other words:

1) every device attached to the BJT base is attached in precisely the

same way to the base terminal of the circuit model.

2) every device attached to the BJT collector is attached in precisely

the same way to the collector terminal of the circuit mode

3) every device attached to the BJT emitter is attached in precisely

the same way to the emitter terminal of the circuit model.

4) every external voltage or current (e.g., vi , vo , iR ) is defined in

precisely the same way both before and after the BJT is replaced
with its circuit model is (e.g., if the output voltage is the collector
voltage in the BJT circuit, then the output voltage is still the
collector voltage in the small-signal circuit!).

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 9/14

It’s just like working in the lab

You can think of replacing a BJT with its small-signal circuit model as a
laboratory operation:

1) Disconnect the red wire (base) of the BJT from the circuit and then
“solder” the red wire (base) of the circuit model to the same point in
the circuit.

2) Disconnect the blue wire (collector) of the BJT from

the circuit and then “solder” the blue wire (collector) of
the circuit model to the same point in the circuit.

3) Disconnect the green wire (emitter) of the BJT from the circuit and then
“solder” the green wire (emitter) of the circuit model to the same point in
the circuit.

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 10/14

This is superposition—
turn off the DC sources!
Step 4: Set all D.C. sources to zero.

Remember:

A zero-voltage DC source is a short circuit.

A zero-current DC source is an open circuit.

The schematic in now in front of you is called the small-signal circuit. Note that
it is missing two things—DC sources and bipolar junction transistors!

* Note that steps three and four are reversible.

You could turn off the DC sources first, and then replace all BJTs with
their small-signal models—the resulting small-signal circuit will be the
same!

* You will find that the small-signal circuit schematic can often be greatly
simplified.

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 11/14

Many things will be connected to ground!

Once the DC voltage sources are turned off, you will find that the terminals of
many devices are connected to ground.

* Remember, all terminals connected to ground are also connected to each

other!

For example, if the emitter terminal is connected to ground, and one

terminal of a resistor is connected to ground, then that resistor terminal is
connected to the emitter!

* As a result, you often find that resistors in different parts of the

circuit are actually connected in parallel, and thus can be combined to
simplify the circuit schematic!

* Finally, note that the AC impedance of a COUS (i.e., ZC = 1 ωC ) is small

for all but the lowest frequencies ω .

If this impedance is smaller than the other circuit elements (e.g., < 10Ω),
we can view the impedance as approximately zero, and thus replace the
large capacitor with a (AC) short!

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 12/14

Organize and simplify or perish!

Organizing and simplifying the small-signal circuit will pay big rewards in the
next step, when we analyze the small-signal circuit.

However, correctly organizing and simplifying the small-signal circuit requires

patience, precision, persistence and professionalism.

Students frequently run into problems when they try to accomplish all the goals
(i.e., replace the BJT with its small-signal model, turn off DC sources, simplify,
organize) in one big step!

Steps 3 and 4 are not rocket science!

Failure to correctly determine the

simplified small-signal circuit is almost
always attributable to an engineer’s
patience, precision and/or persistence
(or, more specifically, the lack of same).

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 13/14

This is a EECS 211 problem,

and only a 211 problem
Step 5: Analyze small-signal circuit.

We now can analyze the small-signal circuit to find all small-signal voltages and
currents.

* For small-signal amplifiers, we typically attempt to find the small-signal

output voltage vo in terms of the small-signal input voltage vi .

From this result, we can find the voltage gain of the amplifier.

* Note that this analysis requires only the knowledge you acquired in EECS
211!

The small-signal circuit will consist entirely of resistors and (small-signal)

voltage/current sources.

These are precisely the same resistors and sources that you learned about
in EECS 211. You analyze them in precisely the same way.

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Steps for Small Signal Analysis lecture 14/14

Trust me, this works!

* Do not attempt to insert any BJT knowledge into your small-signal circuit
analysis—there are no BJTs in a small-signal circuit!!!!!

* Remember, the BJT circuit model contains all of our BJT small-signal
knowledge, we do not—indeed must not—add any more information to the
analysis.

You must trust completely the BJT small-circuit model. It will give
you the correct answer!

Trust the
BJT small-
signal model,
Luke.

Jim Stiles The Univ. of Kansas Dept. of EECS

4/1/2011 Example A Small Signal Analysis of a BJT Amp 1/10

Example: A Small-Signal
Analysis of a BJT
Amplifier
15.0 V
Consider the following BJT
amplifier:

RC =5 K

vO (t ) = VO + vo (t )

RB =5 K
β = 100

vi (t )
+
_

+ RE =5 K COUS
5.8 V
−

Let’s determine its small-signal, open-circuit voltage gain:

vo (t )
Avo =
vi (t )
4/1/2011 Example A Small Signal Analysis of a BJT Amp 2/10

To do this, we must follow each of our five small-signal analysis

steps!

Step 1: Complete a D.C. Analysis

15.0 V
The DC circuit that we must analyze is:

IC
RC =5 K

VO

RB =5 K
β = 100

IB

IE
+ RE =5 K
5.8 V
−

Note what we have done to the original circuit:

1) We turned off the small-signal voltage source

(vi (t ) = 0 ), thus replacing it with a short circuit.

2) We replaced the capacitor with an open circuit—its DC

impedance.
4/1/2011 Example A Small Signal Analysis of a BJT Amp 15.0
3/10 V

Now we proceed with the DC analysis.

IC
We ASSUME that the BJT is in active
RC =5 K
mode, and thus ENFORCE the equalities
VO
VBE = 0.7 V and IC = β IB .
RB =5 K
β = 100
We now begin to +
ANALYZE the circuit by IB
5.8 V
writing the Base-
−
Emitter Leg KVL:
IE
RE =5 K
5.8 − 5IB − 0.7 − 5(β + 1)IB = 0

Therefore:
5.1
IB = = 0.01 mA
5 + 5(101)

and thus:
IC = βIB = 1.0 mA

IE = IB + IC = 1.01 mA

Q: Since we know the DC bias currents, we have all the

information we need to determine the small-signal parameters.

Why don’t we proceed directly to step 2?

4/1/2011 Example A Small Signal Analysis of a BJT Amp 4/10

A: Because we still need to CHECK our assumption! To do this,

we must determine either VCE or VCB .

Note that the Collector voltage is: 15.0 V

VC = 15 − IC RC
= 15 − (1.0)5 IC
RC =5 K
= 10.0 V
VO
And the Emitter voltage is:
RB =5 K
VE = IE RE β = 100
= (1.01)5 +
IB
= 5.05 V 5.8 V
−
Therefore, VCE is: IE
RE =5 K
VCE = VC −VE
= 10.0 − 5.05
= 4.95 V

We now can complete our CHECK:

IC = 1.0 mA > 0

VCE = 4.95 V > 0.7

Time to move on to step 2!

4/1/2011 Example A Small Signal Analysis of a BJT Amp 5/10

Step 2: Calculate the small-signal circuit parameters for

each BJT.

If we use the Hybrid-Π model, we need to determine gm and rπ :

IC 1.0 mA mA
gm = = = 40
VT 0.025V V

VT 0.025 V
rπ = = = 2.5 K
IB 0.01 mA

If we were to use the T-model we would likewise need to

determine the emitter resistance:

VT 0.025 V
re = = = 24.7 Ω
IB 1.01 mA

The Early voltage VA of this BJT is unknown, so we will neglect

the Early effect in our analysis.

As such, we assume that the output resistance is infinite

( ro = ∞ ).
4/1/2011 Example A Small Signal Analysis of a BJT Amp 6/10

Step 3: Carefully replace all BJTs with their small-signal

circuit model. 15.0 V

RC =5 K

RB =5 K B
vO (t ) =
+ C VO + vo (t )
2.5 K
+
vi (t) vbe
_ 40 vbe

+
5.8 V
E
−

RE =5 K COUS
4/1/2011 Example A Small Signal Analysis of a BJT Amp 7/10

Step 4: Set all D.C. sources to zero.

RC =5 K
vo (t )
RB =5 K B

+ C
2.5 K
+
vi (t) vbe
_ 40 vbe

RE =5 K

We likewise notice that the large capacitor (COUS) is an

approximate AC short, and thus we can further simplify the
schematic by replacing it with a short circuit.
4/1/2011 Example A Small Signal Analysis of a BJT Amp 8/10

RC =5 K
vo (t )
RB =5 K B
C
ib + ic
2.5 K
+
vi (t) vbe
_ 40 vbe

ie E

We notice that one terminal of the small-signal voltage source,

the emitter terminal, and one terminal of the collector resistor
RC are all connected to ground—thus they are all collected to
each other!

We can use this fact to simplify the small-signal schematic.

4/1/2011 Example A Small Signal Analysis of a BJT Amp 9/10

RB =5 K B ib ic C vo (t )
+
rπ =
+
vi (t) vbe 2.5 K RC =5 K
40 vbe
_

ie E

The schematic above is the small-signal circuit of this

amplifier. We are ready to continue to step 5!

Step 5: Analyze small-signal circuit.

This is just a simple EECS 211 problem! The left side of the
circuit provides the voltage divider equation:

rπ
vbe = vi
RB + rπ
2.5
= vi
5.0 + 2.5
vi
=
3

a result that relates the input signal to the base-emitter

voltage.
4/1/2011 Example A Small Signal Analysis of a BJT Amp 10/10

RB =5 K B C vo (t )
+
rπ =
+
vi (t) vbe 2.5 K RC =5 K
40 vbe
_

The right side of the schematic allows us to determine the

output voltage in terms of the base-emitter voltage:

vo = −ic RC
= −(gmvbe ) RC
= −40(5)vbe
= −200vbe

Combining these two equations, we find:

vo = −200vbe
vi
= −200
3
= −66.7 vi

The open-circuit, small-signal voltage gain of this amplifier gain

is therefore:

vo
Avo = = −66.7
vi
4/4/2011 Example Small-Signal Input and Output Resistances 1/6

Example: Small-Signal
Input and Output
Resistances
15.0 V
Consider again this circuit:

RC =5 K

vO (t ) = VO + vo (t )

RB =5 K
β = 100

vi (t )
+
_

+ RE =5 K COUS
5.8 V
−

Recall we earlier determined the open-circuit voltage gain Avo of

this amplifier. But, recall also that voltage gain alone is not
sufficient to characterize an amplifier—we likewise require the
amplifier’s input and output resistances!
4/4/2011 Example Small-Signal Input and Output Resistances 2/6

Q: But how do we determine the small-signal input and output

resistances of this BJT amplifier?

A: The same way we always have, only now we apply the

procedures to the small-signal circuit.

Recall that small-signal circuit for this amplifier was

determined to be:

ii (t ) RB =5 K B C vo (t )
+
rπ =
+
_ vi (t) vbe 2.5 K RC =5 K
40 vbe
-

The input resistance of an amplifier is defined as:

vi
Rin =
ii

For this amplifier, it is evident that the input current is:

vi vi vi
ii = = =
RB + rπ 5 + 2.5 7.5

and thus the input resistance of this amplifier is:

4/4/2011 Example Small-Signal Input and Output Resistances 3/6

vi
Rin = = 7.5 K
ii

Now for the output resistance. Recall that determining the

output resistance is much more complex than determining the
input resistance.

The output resistance of an amplifier is the ratio of the

amplifier’s open-circuit output voltage and its short-circuit
output current:

vooc
Rout = sc
io

Again, we determine these values by analyzing the small-signal

amplifier circuit.

First, let’s determine the open-circuit output voltage. This, of

course, is the amplifier output voltage when the output terminal
is open-circuited!

ii (t ) RB =5 K B C vooc (t )
+
rπ =
+
vi (t) vbe 2.5 K RC =5 K
40 vbe
_

E
4/4/2011 Example Small-Signal Input and Output Resistances 4/6

It is evident that the output voltage is simply the voltage across

the collector resistor RC :

vooc = − ( gm vbe ) RC = −40 ( 5 )vbe = −200vbe V

Now, we must determine the short-circuit output current iosc .

This, of course, is the amplifier output current when the output
terminal is short-circuited!

ii (t ) RB =5 K B iosc (t )
C
+
rπ =
vi (t) vbe 2.5 K
+
_
40 vbe RC =5 K
-

It is evident that the short-circuit output current is:

iosc = − gm vbe = −40vbe mA

and therefore the output resistance of this amplifier is:

vooc −200vbe V
Rout = sc = = 5 KΩ
io −40vbe mA

Now we know all three of the parameters required to

characterize this amplifier!
4/4/2011 Example Small-Signal Input and Output Resistances 5/6

Avo = −66.7 V
V

Rin = 7.5 KΩ

Rout = 5.0 KΩ

We can therefore write the equivalent circuit model for this

amplifier:

ii io

5.0K
+ +
+
vi 7.5K vo
−
− −66.7 vi −

Note that the input resistance of this amplifier is not

particularly large, and output resistance is not at all small.

This is not a particularly good voltage amplifier!

4/4/2011 Example Small-Signal Input and Output Resistances 6/6

15.0 V

Now, reflect on how far we have

come. We began with the amplifier RC =5 K
circuit:

vO (t ) = VO + vo (t )

RB =5 K
β = 100

vi (t )
+
_

+ RE =5 K COUS
5.8 V
−

and now we have derived its equivalent small-signal circuit:

ii
vo
5.0K

+
+
vi (t ) 7.5K
_
−
−66.7 vi

With respect to small signal input/output voltages and currents,

these two circuits are precisely the same!