Bipolar Junction Transistors

Emitter p n p Collector Emitter n p n Collector

Base Base
pnp transistor npn transistor

M. B. Patil, IIT Bombay

Bipolar Junction Transistors

Emitter p n p Collector Emitter n p n Collector

Base Base
pnp transistor npn transistor

* Bipolar: both electrons and holes contribute to conduction

* Junction: device includes two p-n junctions (as opposed to a “point-contact” transistor, the first
transistor)
* Transistor: “transfer resistor”
When Bell Labs had an informal contest to name their new invention, one engineer pointed out that it acts like a resistor,
but a resistor where the voltage is transferred across the device to control the resulting current.
([Link]
* invented in 1947 by Shockley, Bardeen, and Brattain at Bell Laboratories.
* BJT is still used extensively, and anyone interested in electronics must have at least a working knowledge
of this device.
* “A BJT is two diodes connected back-to-back.”
WRONG! Let us see why.

M. B. Patil, IIT Bombay

Bipolar Junction Transistors
Consider a pnp BJT in the following circuit:
R1 E C R2
p n p
1k I1 I2 1k
B
I3 10 V
5V

M. B. Patil, IIT Bombay

Bipolar Junction Transistors
Consider a pnp BJT in the following circuit:
R1 E C R2
p n p
1k I1 I2 1k
B
I3 10 V
5V

If the transistor is replaced with two diodes connected back-to-back, we get

R1 E C R2

1k I1 D1 D2 I2 1k
B

5V
I3 10 V

Assuming Von = 0.7 V for D1, we get

5 V 0.7 V
I1 = = 4.3 mA,
R1
I2 = 0 (since D2 is reverse biased), and I3 ⇡ I1 = 4.3 mA.
M. B. Patil, IIT Bombay
Bipolar Junction Transistors

Using a more realistic equivalent circuit for the BJT, we obtain,

R1 E C R2 R1 E ↵ I1 C R2
p n p
1k I1 I2 1k 1k I1 I2 1k
B B
I3 I3
5V 10 V 5V 10 V

M. B. Patil, IIT Bombay

Bipolar Junction Transistors

Using a more realistic equivalent circuit for the BJT, we obtain,

R1 E C R2 R1 E ↵ I1 C R2
p n p
1k I1 I2 1k 1k I1 I2 1k
B B
I3 I3
5V 10 V 5V 10 V

We now get,
5 V 0.7 V
I1 = = 4.3 mA (as before),
R1
I2 = ↵I1 ⇡ 4.3 mA (since ↵ ⇡ 1 for a typical BJT), and
I 3 = I1 I2 = (1 ↵) I1 ⇡ 0 A.
The values of I2 and I3 are dramatically di↵erent than the ones obtained earlier, viz., I2 ⇡ 0, I3 ⇡ 4.3 mA.
Conclusion: A BJT is NOT the same as two diodes connected back-to-back
(although it does have two p-n junctions).

M. B. Patil, IIT Bombay

Bipolar Junction Transistors

What is wrong with the two-diode model of a BJT?

* When we replace a BJT with two diodes, we assume that there is no interaction between the two diodes,
which may be expected if they are “far apart.”

Emitter p n p Collector

Base

Emitter Collector
D1 Base D2

* However, in a BJT, exactly the opposite is true. For a higher performance, the base region is made as
short as possible, and the two diodes cannot be treated as independent devices.

Emitter p n p Collector

Base

* Later, we will look at the “Ebers-Moll model” of a BJT, which is a fairly accurate representation of the
transistor action.

M. B. Patil, IIT Bombay

BJT in active mode

E p n p C E C E n p n C E C
IE IC IE IC IE IC IE IC
IB IB IB
IB
B B B B

* In the active mode of a BJT, the B-E junction is under forward bias, and the B-C junction is under
reverse bias.
- For a pnp transistor, VEB > 0 V , and VCB < 0 V .
- For an npn transistor, VBE > 0 V , and VBC < 0 V .
* Since the B-E junction is under forward bias, the voltage (magnitude) is typically 0.6 to 0.75 V .
* The B-C voltage can be several Volts (or even hundreds of Volts), and is limited by the breakdown voltage
of the B-C junction.
* The symbol for a BJT includes an arrow for the emitter terminal, its direction indicating the current
direction when the transistor is in active mode.
* Analog circuits, including amplifiers, are generally designed to ensure that the BJTs are operating in the
active mode.

M. B. Patil, IIT Bombay

BJT in active mode

E p n p C E C E n p n C E C
IE IC IE IC IE IC IE IC
IB IB IB
IB
B B B B

↵ IE ↵ IE
E C E C
IE IC IE IC
IB IB
B B

M. B. Patil, IIT Bombay

BJT in active mode

E p n p C E C E n p n C E C
IE IC IE IC IE IC IE IC
IB IB IB
IB
B B B B

↵ IE ↵ IE
E C E C
IE IC IE IC
IB IB
B B

* In the active mode, IC = ↵ IE , ↵ ⇡ 1 (slightly less than 1).

* IB = IE IC = IE (1 ↵) .
* The ratio IC /IB is defined as the current gain of the transistor.
IC ↵
= = .
IB 1 ↵
* is a function of IC and temperature. However, we will generally treat it as a constant, a useful
approximation to simplify things and still get a good insight.

M. B. Patil, IIT Bombay

BJT in active mode

E p n p C E C E n p n C E C
IE IC IE IC IE IC IE IC
IB IB IB
IB
B B B B

↵ IE ↵ IE
E C E C
IE IC IE IC
IB IB
B B

IC ↵
= =
IB 1 ↵
* increases substantially as ↵ ! 1.
↵ * Transistors are generally designed to get a high value of
(typically 100 to 250, but can be as high as 2000 for
0.9 9
“super- ” transistors).
0.95 19
* A large ) IB ⌧ IC or IE when the transistor is in the
0.99 99 active mode.
0.995 199

M. B. Patil, IIT Bombay

A simple BJT circuit

1k RC

C 10 V
100 k VCC
= 100
RB B
VBB E
2V
A simple BJT circuit
10 V VCC

10 V VCC 1k RC

1k RC 1k RC IC

C 10 V
VCC n
100 k 100 k p 100 k ↵IE
= 100
RB B VBB RB VBB RB IB
2V n 2V
VBB E
2V IE

Assume the BJT to be in the active mode ) VBE = 0.7 V and IC = ↵IE = IB .
VBB VBE 2 V 0.7 V
IB = = = 13 µA.
RB 100 k
IC = ⇥ IB = 100 ⇥ 13 µA = 1.3 mA.
VC = VCC IC RC = 10 V 1.3 mA ⇥ 1 k = 8.7 V .
Let us check whether our assumption of active mode is correct. We need to check whether the B-C junction is
under reverse bias.
VBC = VB VC = 0.7 V 8.7 V = 8.0 V ,
i.e., the B-C junction is indeed under reverse bias.
M. B. Patil, IIT Bombay
A simple BJT circuit: continued

10 V VCC

10 V VCC 1k RC

1k RC 1k RC IC

C 10 V
VCC n
10 k 10 k p 10 k ↵IE
= 100
RB B VBB RB VBB RB IB
2V n 2V
VBB E
2V IE

What happens if RB is changed from 100 k to 10 k?

M. B. Patil, IIT Bombay

A simple BJT circuit: continued

10 V VCC

10 V VCC 1k RC

1k RC 1k RC IC

C 10 V
VCC n
10 k 10 k p 10 k ↵IE
= 100
RB B VBB RB VBB RB IB
2V n 2V
VBB E
2V IE

What happens if RB is changed from 100 k to 10 k?

Assuming the BJT to be in the active mode again, we have VBE ⇡ 0.7 V , and IC = IB .
VBB VBE 2 V 0.7 V
IB = = = 130 µA ! IC = ⇥ IB = 100 ⇥ 130 µA = 13 mA.
RB 10 k
VC = VCC IC RC = 10 V 13 mA ⇥ 1 k = 3 V
! VBC = VB VC = 0.7 V ( 3) V = 3.7 V .
VBC is not only positive, it is huge!
! The BJT cannot be in the active mode, and we need to take another look at the circuit.
M. B. Patil, IIT Bombay
Ebers-Moll model for a pnp transistor
Active mode (“forward” active mode): B-E in f.b. B-C in r.b.

↵ IE
E p n p C E C E C
IE IC IE IC IE IC
IB IB IB
B B B

Reverse active mode: B-E in r.b. B-C in f.b.

↵R ( IC ) IC
E p n p C E C E C
IE IC IE IC IE IC
IB IB IB
B B B

In the reverse active mode, emitter $ collector. (However, we continue to refer to the terminals with their
original names.)
The two ↵’s, ↵F (forward ↵) and ↵R (reverse ↵) are generally quite di↵erent.
Typically, ↵F > 0.98, and ↵R is in the range from 0.02 to 0.5.
The corresponding current gains ( F and R) di↵er significantly, since = ↵/(1 ↵).
In amplifiers, the BJT is biased in the forward active mode (simply called the “active mode”) in order to make
use of the higher value of in that mode. M. B. Patil, IIT Bombay
Ebers-Moll model for a pnp transistor

The Ebers-Moll model combines the forward and reverse operations of a BJT in a single comprehensive model.

E p n p C I!E ↵F I!E
IE IC
IE D1 IC
IB
E C
B
(p) D2 (p)
E C ↵R I!C
IE IC IB I!C
IB (n) B
B

The currents IE0 and IC0 are given by the Shockley diode equation:
 ✓ ◆  ✓ ◆
VEB VCB
IE0 = IES exp 1 , IC0 = ICS exp 1 .
VT VT

Mode B-E B-C

Forward active forward reverse IE0 IC0
Reverse active reverse forward IC0 IE0
Saturation forward forward IE0 and IC0 are comparable.
Cut-o↵ reverse reverse IE0 and IC0 are negliglbe.

M. B. Patil, IIT Bombay

Ebers-Moll model

pnp transistor

E p n p C I!E ↵F I!E
IE IC
IE D1 IC I!E = IES [exp(VEB /VT ) 1]
IB
E C
B
(p) D2 (p) I!C = ICS [exp(VCB /VT ) 1]
E C
IE IC ↵R I!C I!C
IB
IB (n) B
B

npn transistor
STOP
E n p n C I!E ↵F I!E
IE IC
IE D1 IC I!E = IES [exp(VBE /VT ) 1]
IB
E C
B
(n) D2 (n) I!C = ICS [exp(VBC /VT ) 1]
E C
IE IC ↵R I!C I!C
IB
IB (p) B
B
M. B. Patil, IIT Bombay
Ebers-Moll model in active mode

pnp transistor

E p n p C I!E ↵F I!E
IE IC
IE D1 IC I!E = IES [exp(VEB /VT ) 1]
IB
E C
B
(p) D2 (p) I!C = ICS [exp(VCB /VT ) 1]
E C
IE IC ↵R I!C I!C
IB I C = ↵F I E = F IB
IB (n) B
B

npn transistor

E n p n C I!E ↵F I!E
IE IC
IE D1 IC I!E = IES [exp(VBE /VT ) 1]
IB
E C
B
(n) D2 (n) I!C = ICS [exp(VBC /VT ) 1]
E C
IE IC ↵R I!C I!C
IB I C = ↵F I E = F IB
IB (p) B
B

M. B. Patil, IIT Bombay

BJT I -V characteristics

IC
VCB
p n
B VCE
IB n
VBE IE

M. B. Patil, IIT Bombay

BJT I -V characteristics

IC
VCB
p n
B VCE
IB n
VBE IE

* Since BJT is a three-terminal device, its behaviour can be described in many di↵erent ways, e.g.,
- IC versus VCB for di↵erent values of IE
- IC versus VCE for di↵erent values of VBE
- IC versus VCE for di↵erent values of IB
* The I -V relationship for a BJT is not a single curve but a “family” of curves or “characteristics.”
* The IC -VCE characteristics for di↵erent IB values are useful in understanding amplifier biasing.

M. B. Patil, IIT Bombay

 BJT I -V characteristics

VCE
E C
IE IC
IB
B
IB0
10 µA

↵F
↵F = 0.99 ! F = = 99
1 ↵F
↵R
↵R = 0.5 ! R = =1
1 ↵R
14
IES = 1 ⇥ 10 A
14
ICS = 2 ⇥ 10 A
 BJT I -V characteristics
1.0

VCE 0.5 VBE (Volts)

I!E 0.0
↵F I!E
VCE −0.5 VBC (Volts)
IE D1 IC
E C
E C −1.0
IE IC (n) D2 (n)
−1.5
IB ↵R I!C 20
B IB I!C
IB0 (p) B I!C (µA)
10 µA IB0 10
10 µA
↵F
↵F = 0.99 ! F = = 99
1 ↵F
↵R I!E = IES [exp(VBE /VT ) 1] 0
↵R = 0.5 ! R = =1
1 ↵R 1.2
14
I!C = ICS [exp(VBC /VT ) 1] sat lin
IES = 1 ⇥ 10 A
ICS = 2 ⇥ 10 14
A IC = ↵F IE = F IB in active mode 0.8

0.4
* linear region: B-E under forward bias, B-C under reverse bias, IC = F IB IC (mA)
* saturation region: B-E under forward bias, B-C under forward bias, IC < F IB 0
I!E (mA)

VCE :0 0.5 1 1.5 2

M. B. Patil, IIT Bombay
 BJT I -V characteristics

1.0
VCE
0.5 VBE (Volts)
I!E ↵F I!E
VCE 0.0
IE D1 IC
E C VBC (Volts)
E
IE IC (n) D2 C −0.5
(n)
IB ↵R I!C
B IB I!C −1.0
10 µA (p) B
20 µA −1.5
10 µA
20 µA sat lin
2
↵F IB = 20 µA
↵F = 0.99 ! F = = 99
1 ↵F
↵R I!E = IES [exp(VBE /VT ) 1]
↵R = 0.5 ! R = =1
1 ↵R
I!C = ICS [exp(VBC /VT ) 1]
IES = 1 ⇥ 10−14 A 1
−14 IC = ↵F IE = F IB in active mode
IB = 10 µA
ICS = 2 ⇥ 10 A

* linear region: B-E under forward bias, B-C under reverse bias, IC = F IB IC (mA)
0
* saturation region: B-E under forward bias, B-C under forward bias, IC < F IB 0 0.5 1 1.5 2
VCE (Volts)

M. B. Patil, IIT Bombay

A simple BJT circuit (revisited)
saturation

linear
10 V VCC 15

1k RC IB = 130 µA (RB = 10 k)
10
= 100 IC

IC (mA)
n load line
p
5
VBB RB I B n
2V
IE IB = 13 µA (RB = 100 k)

0
0 2 4 6 8 10
VCE (V)

We are now in a position to explain what happens when RB is decreased from 100 k to 10 k in the above circuit.
VBB 0.7 V
Let us plot IC VCE curves for IB ⇡ for the two values of RB .
RB
In addition to the BJT IC VCE curve, the circuit variables must also satisfy the constraint,
VCC = VCE + IC RC , a straight line in the IC VCE plane.
The intersection of the load line and the BJT characteristics gives the solution for the circuit. For RB = 10 k,
note that the BJT operates in the saturation region, leading to VCE ⇡ 0.2 V , and IC = 9.8 mA.
M. B. Patil, IIT Bombay
BJT circuit example

Assuming the transistor to be operating in the active region, find RE and RC to obtain IE = 2 mA, and
VBC = 1 V (↵ ⇡ 1).

IE E C IC

RE B RC
5V 5V

VEE VCC

M. B. Patil, IIT Bombay

BJT circuit example

Assuming the transistor to be operating in the active region, find RE and RC to obtain IE = 2 mA, and
VBC = 1 V (↵ ⇡ 1).

IE E C IC

RE B RC
5V 5V

VEE VCC

4.3 V
VEB VEE + IE RE = 0 ! IE R E = 5 0.7 ! RE = = 2.15 k.
2 mA
VBC + IC RC VCC = 0 ! IC RC = VCC VBC .
4V
Since ↵ ⇡ 1, IC ⇡ IE ! IE R C ⇡ 5 1 ! RC = = 2 k.
2 mA

M. B. Patil, IIT Bombay