Frequency Response of BJT

The Mid-frequency small-signal models

ib ic
b c
+ + +
vbe r v rd vce
gmv
_ _ _
e e
hybrid- model

38.92
gm = I C (Note: Uses DC value of I C )
n
where n = 1 (typical, Si BJT)
1
 o = h fe rd =
h oe
o
h re = 0 r = h ie =
gm
 ib ic
b c
+ +
vbe re vce
ib
_ _
e e
re model

26 mV
re = (Note: uses DC value of I B )
IB
 o = h fe
 o re = h ie
h re = 0
1
h oe = 0, or use rd =
h oe
 BJT Internal Capacitances
The Collector-Base Junction capacitance:
In active mode collector base junction is reverse biased and its
depletion capacitance
Value of Cµ at zero
C voltage
0
C m
VCB
1 V
0c

Grading coefficient
CBJ Built in of EBJ (0.2 to 0.5)
voltage
(0.75V)
 BJT Internal Capacitances
۞Transistor exhibit charge storage phenomenon that limit the
speed and frequency of their operation
۞These charge effects are accounted by adding capacitances
to the hybrid π model.
Base charging or diffusion capacitance Cde
۞When the transistor is operating in active or Saturation region
mode, minority carrier charge is stored in the base region.
۞When an npn transistor is operating in active mode, this
charge Qn is represented by
Qn i
F C Wher F W 2
e 2Dn
F is called as Forward base transient time
dQn diC IC
C de dv BE F
dv BE F gm F
VT
 BJT Internal Capacitances
The Base Emitter Junction Capacitance:
(depletion layer capacitance)
Value of Cje at zero
voltage

Approximate value C je
of Cje =2 Cje0 C je m
0 VBE
1 V
0e

Grading coefficient
EBJ Built in voltage of EBJ (0.5)
(0.9)
 High frequency Hybrid-π
model
Emitter base capacitance (Cde+Cje) pfarads
Collector base capacitance
de

je

B
Model resistance of
silicon material of
the base region ,
between base B
and a fictitious
internal or intrinsic,
base terminal B’
(right under the
emitter region)
 High frequency Hybrid-π
model
Emitter base capacitance (Cde+Cje)pfarads Collector
base
B capacitance
B’

(Few tens of Ωs)

Since rx <<< rπ , its
effect is negligible at
low frequencies , it
becomes apparent in
the high frequencies.
 Cut off frequency
۞Value of Cπ is not given in the data sheet rather behavior
of β (or hfe) versus frequency is normally given.
۞To find Cπ and Cµ, derive an expression for hfe. i.e short
circuit current gain as a function of frequency in terms of
hybrid π components.
Consider following model
 Cut off frequency
Collector is shorted to the emitter at Junction C

E
 Cut off frequency

Ib sCµVπ Ic=(gm-sCµ)Vπ

E Ic sC V gmV
Ic gmV sC V (gm sC )V
 Cut off frequency
Ib sCµVπ Ic=(gm-sCµ)Vπ

Vπ = Ib x [Impedance sum E V Ib [r || C || C ]
between B’ and E] Ib
1
sC sC
r
 Cut off frequency
Ic gm sC
h fe (or )
Ib 1
s(C C )
r
The model is valid for gm >>> wCµ , we can neglect wCµ in the
numerator
Ic gmr 0
hfe (or )
Ib 1 s(C C )r 1 s(C C )r
Low frequency value of beta This has a single pole response
with
3dB frequency at w=wβ
1
Where
w
(C C )r
 Cut off frequency

We can observe from the above slope that frequency at which

|hfe| drops to unity is called as unity gain bandwidth wT and is
given by wT=β0wβ
fT is some times specified in the
gm gm datashet or given as a function of
wT ; fT
(C C ) 2 (C C ) Ic and CE
V