FREQUENCY MODULATION (ADC-UNIT-II)

There is another method of modulating a sinusoidal carrier wave, namely, angle Modulation in
which either the phase or frequency of the carrier wave is varied according to the message signal.
Thus there two types of Angle modulation techniques namely
1. Phase modulation
2. Frequency modulation
Phase Modulation (PM)
This is one of the angle modulation techniques in which the phase of the high frequency sine
wave carrier is varied directly by the message signal. Let the carrier be
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆[2𝜋𝑓𝐶 𝑡 + ∅(𝑡)]
Where 𝐴𝐶 is constant amplitude, fc is the carrier frequency. ∅(𝑡)is the phase of the carrier wave
which is according to the definition
∅(𝑡) ∝ 𝑚(𝑡)
∴ ∅(𝑡) = 𝑘𝑝 𝑚(𝑡)
Where 𝐴𝐶 is constant amplitude, fc is the carrier frequency, m(t) is the message signal and 𝑘𝑝 is
called as phase sensitivity, now the equation of PM is given by
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆[2𝜋𝑓𝐶 𝑡 + 𝑘𝑝 𝑚(𝑡)]

Frequency Modulation (FM)

This is another type of angle modulation in which the frequency of the carrier (not the phase as
in PM) is varied directly with the message signal. Let the carrier be
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆[2𝜋𝑓𝑖 𝑡]
Where 𝐴𝐶 is constant amplitude and 𝑓𝑖 instantaneous carrier frequency and is according the
definition as
𝑓𝑖 ∝ 𝑚(𝑡)when𝑚(𝑡) > 0
∴ 𝑓𝑖 = 𝑘𝑓 𝑚(𝑡)when𝑚(𝑡) > 0 − − − − − − − − − − − (1)
𝑤ℎ𝑒𝑟𝑒 𝑘𝑓 𝑖𝑠 𝑎 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 𝑜𝑓 𝑝𝑟𝑜𝑝𝑜𝑟𝑡𝑖𝑜𝑛𝑎𝑙𝑖𝑡𝑦 𝑎𝑛𝑑 𝑐𝑎𝑙𝑙𝑒𝑑 𝑎𝑠 𝑓𝑟𝑒𝑞𝑒𝑢𝑛𝑐𝑦 𝑠𝑒𝑛𝑠𝑖𝑡𝑖𝑣𝑖𝑡𝑦
And also 𝑓𝑖 = 𝑓𝐶 when𝑚(𝑡) = 0 − − − − − − − − − − − (2)
Therefore
∴ 𝑓𝑖 = 𝑓𝐶 + 𝑘𝑓 𝑚(𝑡) − − − − − − − − − − − −(3)
Multiply on both side of eq(3) 2 and then integrate
2𝜋𝑓𝑖 = 2𝜋𝑓𝐶 + 2𝜋𝑘𝑓 𝑚(𝑡)
𝜔𝑖 = 2𝜋𝑓𝐶 + 2𝜋𝑘𝑓 𝑚(𝑡)
𝑡
𝑡 𝑡
𝑑𝜃
∫ 𝑑𝑡 = ∫ 2𝜋𝑓𝐶 𝑑𝑡 + ∫ 2𝜋𝑘𝑓 𝑚(𝑡) 𝑑𝑡
𝑡 𝑑𝑡 0
0
𝑡

𝜃𝑖 = 2𝜋𝑓𝐶 𝑡 + 2𝜋𝑘𝑓 ∫ 𝑚(𝑡) 𝑑𝑡

0
Therefor the equation of FM signal is given by
𝑡
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆 [2𝜋𝑓𝐶 𝑡 + 2𝜋𝑘𝑓 ∫0 𝑚(𝑡) 𝑑𝑡]
There is a close relationship between PM and FM. From the phase and frequency relations of PM and FM
𝑡
given above it is clear that replacing m(t) in the expression of PM signal with∫0 𝑚(𝑡) 𝑑𝑡 we can get the
expression of FM except scaling factor. Similarly the expression of FM can be obtained by replacing
𝑡
∫0 𝑚(𝑡) 𝑑𝑡 by m(t). This is illustrated in the following block diagram.

m(t) Integrator Phase Modulator FM

(PM)

m(t) Differentiator Frequency Modulator

PM
(PM)
Single tone Frequency Modulation:
In Frequency Modulation (FM) the instantaneous frequency changes according to the
instantaneous value of the information signal controls the frequency of the carrier wave. This
process is illustrated in terms of waveforms in the following diagrams.

Notice that as the information signal increases, the frequency of the carrier increases, and as the
information signal decreases, the frequency of the carrier decreases. The frequency fi of the
information signal controls the rate at which the carrier frequency increases and decreases. As
with AM, fi must be less than fc. The amplitude of the carrier remains constant throughout this
process.
When the information voltage reaches its maximum value then the change in frequency of the
carrier will have also reached its maximum deviation above the nominal value. Similarly when
the information reaches a minimum the carrier will be at its lowest frequency below the nominal
carrier frequency value. When the information signal is zero, then no deviation of the carrier will
occur. The maximum change that can occur to the carrier from its base value fc is called the
frequency deviation, and is given the symbol fc. This sets the dynamic range (i.e. voltage range)
of the transmission. The dynamic range is the ratio of the largest and smallest analogue
information signals that can be transmitted.
Time Domain Representation of Single Tone FM:
The expression of FM signal is given by
𝑡

𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆 [2𝜋𝑓𝐶 𝑡 + 2𝜋𝑘𝑓 ∫ 𝑚(𝑡) 𝑑𝑡 ]

0
For single tone 𝑚(𝑡) = 𝐴𝑚 𝐶𝑂𝑆[2𝜋𝑓𝑚 𝑡]
Therefore FM is
𝑡

𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆 [2𝜋𝑓𝐶 𝑡 + 2𝜋𝑘𝑓 ∫ 𝐴𝑚 𝐶𝑂𝑆[2𝜋𝑓𝑚 𝑡] 𝑑𝑡]

0
2𝜋𝑘𝑓 𝐴𝑚
= 𝐴𝐶 𝐶𝑂𝑆 [2𝜋𝑓𝐶 𝑡 + 𝑠𝑖𝑛[2𝜋𝑓𝑚 𝑡]]
2𝜋𝑓𝑚
𝐴𝑚 𝑘𝑓
= 𝐴𝐶 𝐶𝑂𝑆 [2𝜋𝑓𝐶 𝑡 + 𝑠𝑖𝑛[2𝜋𝑓𝑚 𝑡]]
𝑓𝑚
∆𝑓
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆 [2𝜋𝑓𝐶 𝑡 + 𝑠𝑖𝑛[2𝜋𝑓𝑚 𝑡]]
𝑓𝑚
Where ∆f is the frequency deviation which is defined as the maximum departure of the frequency
of FM from its unmodulated carrier frequency and is equal to Am k f
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆[2𝜋𝑓𝐶 𝑡 + 𝛽sin(2𝜋𝑓𝑚 𝑡)]
∆f
Where β is called modulation index and equal to β = f . If value of β is less than one then the
m
FM signal is known as Narrow Band Frequency Modulation (NBFM) otherwise if β is greater
than one the cooresponding FM is known as Wide Band Frequency Modulation (WBFM).

Narrow Band Frequency Modulation (NBFM) – Single Tone

The single tone FM signal is given by
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆[2𝜋𝑓𝐶 𝑡 + 𝛽sin(2𝜋𝑓𝑚 𝑡)]
Expand the above equation
𝑠(𝑡) = 𝐴𝐶 cos[2𝜋𝑓𝐶 𝑡] cos[𝛽 𝑠𝑖𝑛(2𝜋𝑓𝑚 𝑡)] + 𝐴𝐶 sin[2𝜋𝑓𝐶 𝑡] sin[𝛽 sin(2𝜋𝑓𝑚 𝑡)]
If 𝛽 cos[𝛽 𝑠𝑖𝑛(2𝜋𝑓𝑚 𝑡)] = 1 and sin[𝛽 sin(2𝜋𝑓𝑚 𝑡)] = 𝛽 sin(2𝜋𝑓𝑚 𝑡)
𝑠(𝑡) = 𝐴𝐶 cos(2𝜋𝑓𝐶 𝑡) + 𝐴𝐶 𝛽sin(2𝜋𝑓𝐶 𝑡) sin(2𝜋𝑓𝑚 𝑡) -------------------- (1)
𝐴𝐶 𝛽
𝑆(𝑓) = 𝐴𝐶 cos(2𝜋𝑓𝐶 𝑡) + {cos[2𝜋(𝑓𝐶 + 𝑓𝑚 )𝑡] − cos[2𝜋(𝑓𝐶 − 𝑓𝑚 )𝑡] }
2
Wide Band Frequency Modulation (WBFM) – Single Tone:
The single tone FM signal is given by
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆[2𝜋𝑓𝐶 𝑡 + 𝛽sin(2𝜋𝑓𝑚 𝑡)]
The above equation can be written as
𝑠(𝑡) = 𝑅𝑒[𝐴𝐶 𝑒 𝑗[2𝜋𝑓𝐶 𝑡+𝛽 sin(2𝜋𝑓𝑚𝑡)] ]
= 𝑅𝑒[𝐴𝐶 𝑒 𝑗[2𝜋𝑓𝐶 𝑡] 𝑒 𝑗[𝛽 sin(2𝜋𝑓𝑚 𝑡)] ]
= 𝑅𝑒[𝑠̃ (𝑡)𝑒 𝑗[2𝜋𝑓𝐶 𝑡] ]
Where 𝑠̃ (𝑡) = 𝐴𝐶 𝑒 𝑗[𝛽 sin(2𝜋𝑓𝑚𝑡)]
Since sin(2nƒnt) is periodic with fundamental period T = 1/fn , the complex exponential
𝑒 𝑗[𝛽 sin(2𝜋𝑓𝑚𝑡)] is also periodic with the same fundamental period. Therefore, this complex
exponential can be expanded in Fourier series representation as
∝
𝑗[𝛽 sin(2𝜋𝑓𝑚 𝑡)]
𝑒 = ∑ 𝐶𝑛 𝑒 𝑗2𝜋𝑛𝑓𝑚𝑡
𝑛=−∝

Where 𝐶𝑛 the Fourier series coefficient and is given by

∝ ∝
𝐶𝑛 = ∫−∝ 𝑒 𝑗[𝛽 sin(2𝜋𝑓𝑚𝑡)] 𝑒 −𝑗2𝜋𝑛𝑓𝑚 𝑡 𝑑𝑡 = ∫−∝ 𝑒 𝑗[𝛽 sin(2𝜋𝑓𝑚 𝑡)−2𝜋𝑛𝑓𝑚 𝑡] 𝑑𝑡
Let 2𝜋𝑓𝑚 𝑡 = 𝑥 Then
𝜋
1
𝐶𝑛 = ∫ 𝑒𝑗[𝛽 sin(𝑥)−𝑛𝑥] 𝑑𝑥 = 𝐽𝑛 (𝛽)
2𝜋
−𝜋
Therefore the integral on the right side is known as the nth order Bessel function of the first kind
and is denoted as Jn (β). Therefore ej[β sin(2πfm t)] can be written as
∝

𝑒 𝑗[𝛽 sin(2𝜋𝑓𝑚𝑡)] = ∑ 𝐽𝑛 (𝛽) 𝑒 𝑗2𝜋𝑛𝑓𝑚 𝑡

𝑛=−∝

Now
∝
𝑗[𝛽 sin(2𝜋𝑓𝑚 𝑡)]
𝑠̃ (𝑡) = 𝐴𝐶 𝑒 = 𝐴𝐶 ∑ 𝐽𝑛 (𝛽) 𝑒 𝑗2𝜋𝑛𝑓𝑚𝑡
𝑛=−∝
𝑠(𝑡) = 𝑅𝑒[𝑠̃ (𝑡)𝑒 𝑗[2𝜋𝑓𝐶 𝑡] ]
= 𝑅𝑒[𝐴𝐶 ∑∝𝑛=−∝ 𝐽𝑛 (𝛽) 𝑒𝑗2𝜋𝑛𝑓𝑚𝑡 𝑒 𝑗[2𝜋𝑓𝐶 𝑡] ]
= 𝑅𝑒[𝐴𝐶 ∑∝𝑛=−∝ 𝐽𝑛 (𝛽) 𝑒𝑗2𝜋(𝑓𝐶+𝑛𝑓𝑚)𝑡 ]
∝

= 𝐴𝐶 ∑ 𝐽𝑛 (𝛽)𝐶𝑂𝑆[( 𝑓𝐶 + 𝑛𝑓𝑚 )𝑡]

𝑛=−∝

The spectrum is obtained by taking Fourier transform on both sides

∝
𝐴𝐶
𝑆(𝑓) = ∑ 𝐽𝑛 (𝛽) [ 𝛿( 𝑓 − ̅̅̅̅̅̅̅̅̅̅̅̅
𝑓𝐶 + 𝑛𝑓𝑚 ) + 𝛿(𝑓 + ̅̅̅̅̅̅̅̅̅̅̅̅
𝑓𝐶 + 𝑛𝑓𝑚 )]
2
𝑛=−∝
Generation of FM waves:
FM waves are normally generated by two methods. They are 1) Indirect method and 2) Direct
method. In both methods first FM is generated as NBFM later it is converted into WBFM by
using frequency multipliers.
Indirect Method (Armstrong Method) of FM Generation
In this method first NBFM is generated first by using phase modulator and then the wideband
FM with desired frequency deviation (and modulation index) is obtained by using frequency
multipliers. The equation for FM wave is given by
For the narrow band FM wave ∅(𝑡) is small then

-------------------- (2)
Narrow band FM is generated by implementing the above equation (1) or (2) as shown in the
figure below.
NBFM

− Frequency WBFM
m(t) Integrator
Product
Modulator + Multiplier
+

90o Phase Carrier

Shifter Generator

Narrow Band PM
The frequency deviation ∆ƒ is very small in narrow-band FM wave. To produce wideband FM,
we have to increase the value of ∆ƒ to a desired level. This is achieved by means of one or
multiple frequency multipliers. A frequency multiplier consists of a nonlinear device and a band
pass filter. Then the order of nonlinear device produces a dc component and number of
frequency modulated waves with carrier frequencies ƒc, 2ƒc,…nƒc and frequency deviations ∆ƒ,
2∆ƒ,…n∆ƒ, respectively. If we want an FM wave with frequency deviation of 6∆ƒ, then we may
use a 6th order nonlinear device or one 2nd order and one 3rd order nonlinear followed by a band
pass filter centered at 6ƒc. Normally, we may require very high value of frequency deviation.
This automatically increases the carrier frequency by the same factor which may be higher than
the required carrier frequency. We may shift the carrier frequency to the desired level by using
mixer which does not change the frequency deviation.
Direct Method of FM Generation
In this method, the instantaneous frequency ƒ(t) of the carrier signal c(t) is varied directly with
the instantaneous value of the modulating signal n(t). For this, an oscillator is used in which
anyone of the reactive components (either C or L) of the resonant network of the oscillator is
varied linearly with m(t). We can use a varactor diode or a varicapas as a voltage-variable
capacitor whose capacitance solely depends on the reverse-bias voltage applied across it. To
vary such capacitance linearly with m(t), we have to reverse-bias the diode by the fixed DC
voltage and operate within a small linear portion of the capacitance-voltage characteristic curve.
The unmodulated fixed capacitance C0 is linearly varied by m(t) such that the resultant
capacitance becomes
𝐶(𝑡) = 𝐶𝑜 + 𝑘𝐶 𝑚(𝑡)
Thus the voltage controlled oscillator (VCO) acts an FM generator. The above discussed
scheme can be implemented in Hartley oscillator as shown in the figure.

The frequency of oscillation for such an oscillator is given by

1 1
𝑓𝑖 = =
2𝜋√(𝐿1 + 𝐿2 )𝐶(𝑡) 2𝜋√(𝐿1 + 𝐿2 )[𝐶𝑜 + 𝑘𝐶 𝑚(𝑡)]
1
=
𝑘𝐶
2𝜋√(𝐿1 + 𝐿2 )𝐶𝑜 [1 + 𝑚(𝑡)]
𝐶𝑜

𝑓𝑜
=
𝐶 𝑘
√[1 + 𝐶 𝑚(𝑡)]
𝑜

Where fo center frequency and equal to

1
𝑓𝑜 =
2𝜋√(𝐿1 + 𝐿2 )𝐶𝑜
1
−
𝑘𝐶 2 𝑘𝐶 𝑓 𝑜 𝑘𝐶
𝑓𝑖 = 𝑓𝑜 [1 + 𝑚(𝑡)] = 𝑓𝑜 [1 − 𝑚(𝑡) + ⋯ ] = [𝑓𝑜 − 𝑚(𝑡) + ⋯ ]
𝐶𝑜 2𝐶𝑜 2𝐶𝑜
kC
When |2Co m(t)| ≪ 1 higher order terms in the above expansion can be neglected. Therefore
𝑓𝑖 = 𝑓𝑜 + 𝑘𝑓 𝑚(𝑡)
𝑓𝑜 𝑘𝐶
𝑊ℎ𝑒𝑟𝑒 𝑘𝑓 = − = 𝑓𝑟𝑒𝑞𝑢𝑒𝑛𝑐𝑦 𝑠𝑒𝑛𝑠𝑖𝑡𝑖𝑣𝑖𝑡𝑦
2𝐶𝑜
Thus the instantaneous frequency fi is linearly varies with message signal m(t). The FM that we
obtained has low frequency deviation (so it is NBFM) hence the frequency deviation has to be
increased through the frequency multiplication as shown in the diagram below to get WBFM.
 m(t) VCO n1 Mixer n2 WBFM
M

Crystal
Oscillator

NBFM Frequency
multiplier
WBFM
M
Transmission bandwidth of FM wave
The transmission bandwidth of an FM wave depends on the modulation index β. The
modulation index, on the other hand, depends on the modulating amplitude and modulating
frequency. It is almost impossible to determine the exact bandwidth of the FM wave. Rather,
we use a rule-of-thumb expression for determining the FM bandwidth.

For single-tone frequency modulation, the approximated bandwidth is determined by the

expression

This expression is regarded as the Carson’s rule. The FM bandwidth determined by this rule
can accommodate at least 98% of the total power. For an arbitrary message signal m(t) with
bandwidth or maximum frequency W, the Bandwidth of the corresponding FM wave may be
determined by Carson’s rule as

Pre-emphasis and De-emphasis

The Signal to Noise ratio (SNR) decreases with increases in frequency as the baseband signals
are relatively rich in lower frequencies than higher frequencies. In communication systems the
system performance depends of SNR. As a result of this, the communication system response is
more for lower frequencies than higher frequencies. Thus the higher components of message
signal are badly affected by the noise. To solve this problem, we can use a pre-emphasis filter
of transfer function Hp(ƒ) at the transmitter to boost the higher frequency components before
modulation. Similarly, at the receiver, the de-emphasis filter of transfer function Hd(ƒ) can be
used after demodulator to attenuate the higher frequency components there by restoring the
original message signal.
Pre-emphasis: It is the process of boosting up the amplitude of higher frequencies relatively
with that of low frequencies according to a predetermined curve as shown in the plot below.
De-emphasis: It is the process of attenuating the higher frequency components relatively with
low frequency components according to a predetermined curve as shown in the plot below.
 + dB
Pr-emphasis

2.123 Khz

15 Khz f

De-emphasis

- dB

Balanced Frequency Discriminator

A frequency discriminator consists of a slope circuit followed by an envelope detector. The
transfer function of single frequency discriminator is nonlinear to some extent. On other hand a
balanced frequency discriminator consists of two such frequency discriminators to have a linear
transfer function. The block diagram and circuit diagram of balanced frequency discriminator is
shown in the figure below. The frequency response also is shown in the figure.

The slope circuit converts an FM signal into a hybrid modulated signal consisting of amplitude
variations along with frequency variation corresponding to the message signal. The ideal slope
circuit is characterized by a transfer function that is purely imaginary and varies linearly with
frequency inside the prescribed frequency interval as shown in the figure below.
 𝐻(𝑓)
𝑗

Slope = 2πa
𝐵𝑇 𝐵𝑇
−𝑓𝑐 − −𝑓𝑐 +
2 2 𝑓
𝐵𝑇 𝐵𝑇
𝑓𝑐 − 𝑓𝑐 +
2 2

The above transfer function can be represented mathematically

̅̅̅̅̅̅̅̅̅̅
𝐵𝑇 𝐵𝑇 𝐵𝑇
𝐻(𝑓) = 𝑗2𝜋𝑎 (𝑓 − 𝑓𝑐 − ) 𝑓𝑜𝑟 𝑓𝑐 − ≤ 𝑓 ≤ 𝑓𝑐 +
2 2 2
̅̅̅̅̅̅̅̅̅̅
𝐵𝑇 𝐵𝑇 𝐵𝑇
𝐻(𝑓) = 𝑗2𝜋𝑎 (𝑓 + 𝑓𝑐 − ) 𝑓𝑜𝑟 −𝑓𝑐 − ≤ 𝑓 ≤ −𝑓𝑐 +
2 2 2

Let the 𝐻̃1 (𝑓) be the complex transfer function of 𝐻1 (𝑓) of the upper slope circuit and is given
by
̃1 (𝑓) = 𝐻1 (𝑓 + 𝑓𝑐 ) = 𝑗2𝜋𝑎 ( 𝑓 + 𝐵𝑇 )
𝐻 −
𝐵𝑇
≤𝑓≤
𝐵𝑇
2 2 2
The incoming FM wave is given by
𝑡
𝑠(𝑡) = 𝐴𝐶 𝐶𝑂𝑆 [2𝜋𝑓𝐶 𝑡 + 2𝜋𝑘𝑓 ∫0 𝑚(𝑡) 𝑑𝑡] = 𝑅𝑒[𝑠̃ (𝑡)𝑒 𝑗[2𝜋𝑓𝐶 𝑡]
𝑡
𝑚(𝑡) 𝑑𝑡
Where 𝑠̃ (𝑡) = 𝐴𝑐 𝑒 𝑓 ∫0
𝑗2𝜋𝑘

Let s̃1 (t) be the envelope of the response of the slope circuit. Then the Fourier transform of s̃1 (t)
is 𝑆̃1 (𝑓)
̃1 (𝑓)𝑆̃(𝑓) = 𝑗2𝜋𝑎 ( 𝑓 + 𝐵𝑇 ) 𝑆̃(𝑓)
𝑆̃1 (𝑓) = 𝐻
2
Therefore
𝑑[s̃1 (t)]
s̃1 (t) = 𝑎 [ + 𝑗𝜋𝑎𝐵𝑇 s̃1 (t)]
𝑑𝑡

2𝑘𝑓 𝑡
𝑗2𝜋𝑘𝑓 ∫0 𝑚(𝑡) 𝑑𝑡
𝑠̃1 (𝑡) = 𝑗𝜋𝑎𝐵𝑇 𝐴𝐶 [1 + 𝑚(𝑡)] 𝑒
𝐵𝑇
2𝑘𝑓 𝑡 𝜋
𝑠1 (𝑡) = 𝑅𝑒[𝑠̃ (𝑡)𝑒 𝑗[2𝜋𝑓𝐶 𝑡] = 𝜋𝑎𝐵𝑇 𝐴𝐶 [1 + 𝑚(𝑡)] 𝐶𝑂𝑆 [2𝜋𝑓𝐶 𝑡 + 2𝜋𝑘𝑓 ∫0 𝑚(𝑡) 𝑑𝑡 + ]
𝐵𝑇 2
Therefore the envelope of the signal 𝑠1 (𝑡) at the output of the upper envelope detector is given by
2𝑘𝑓
|𝑠1 (𝑡)| = 𝜋𝑎𝐵𝑇 𝐴𝐶 [1 + 𝑚(𝑡)]
𝐵𝑇

Similarly the complex transfer function of the lower slope circuit is assumed to be
̃2 (𝑓) = 𝐻
𝐻 ̃1 (−𝑓). Hence the envelope of the signal 𝑠2 (𝑡) at the output of the lower envelope
2𝑘
detector is given by |𝑠2 (𝑡)| = 𝜋𝑎𝐵𝑇 𝐴𝐶 [1 − 𝐵 𝑓 𝑚(𝑡)]
𝑇
Finally the output of FM detector is
|𝑠0 (𝑡)| = |𝑠1 (𝑡)| − |𝑠2 (𝑡)| = 4𝜋𝑎𝐴𝐶 𝑘𝑓 𝑚(𝑡)
Thus the output is proportional to the message signal except the scaling factor 4𝜋𝑎𝐴𝐶 𝑘𝑓
Ratio detector
The ratio detector is a variant of the frequency discriminator. The circuit is similar to the
discriminator, but in a ratio detector, the diodes conduct are arranged to conduct in opposite
directions and the primary voltage added to the center tap of the secondary of the input
transformer. Also, the output is not taken across the diodes as in frequency discriminator, but
between the junction of the capacitors and resistors as shown in the diagram. The output across
the diodes (as well as resistors) is connected to a large capacitor, which eliminates AM noise in
the ratio detector output. The operation of the ratio detector is very similar to the discriminator,
but the output is only 50% of the output of a discriminator for the same input signal.
In the circuit diagram the capacitor Co charges to a voltage V = (V1 + V2). Normally the
R1 and R2 are equal and the voltage at the junction of R1 and R2 measured with respect to the
bottom point equals (V1 + V2)/2. This voltage remains fixed because of large capacitor Co and
large time constant.

The given input FM will be converted to hybrid modulated signal having both amplitude
variations and frequency variations by the slope circuits formed from the secondary winding of
the input transformer. These voltages are represented as Vs1 and Vs2 respectively. These
secondary voltages added to the primary voltage with a phase shift of 90o and the resulting
combined voltage is applied to the respective envelop detectors. The output of the upper
envelope detector (V1) and lower envelope detectors (V2) are added and charges the capacitor
Co to (V1 + V2). The demodulated output is measured between Va and Vb. Therefore
VCo = V = V1 + V2
V1 + V2 V1 − V2
Vo = Va − Vb = − V2 =
2 2
The response of the detector to different frequencies of FM operation is discussed in three
different cases as follows with the help of phasor diagrams.
Case-1: For FM signal frequency equal to carrier frequency ( f = fc )
When f = fc , the loop current in secondary winding of the input transformer, Is, is in phase with
the secondary voltage, Vs,. Hence the half secondary voltage across the upper part of the
secondary, Vs1, is lagging Is by 90o. Half secondary voltage across the lower part of the
secondary, Vs2, is lagging Vs1 by 180o as the terminal voltages of secondary winding of a
transformer differ by 180o. The vector sum of Vs1 and Vp is V1. The vector sum of Vs2 and Vp
is V2. Therefore from the diagram

Case-1 (f = fc) Case-2 (f > fc) Case-3 (f < fc)

V1 Vs1 V1 V1
Vs1 Vs1

Is Is
Vs Vp Vs Vp Vs Vp
Is
Vs2 Vs2
V2 V2 Vs2 V2

Case-1 Case-2
𝑉1 = 𝑉2 𝑉 𝑉
𝑉1 > 𝑉2 , 𝑉1 = 2 + ∆𝑉 And 𝑉2 = 2 − ∆𝑉
𝑇ℎ𝑒𝑟𝑒𝑓𝑜𝑟𝑒
𝑇ℎ𝑒𝑟𝑒𝑓𝑜𝑟𝑒
𝑉𝐶0 = 𝑉1 + 𝑉1 = 2𝑉1 = 2𝑉2 𝑉 𝑉
V1 − V2 𝑉𝐶0 = 𝑉1 + 𝑉2 = + ∆𝑉 + 2 − ∆𝑉 = V and
2
𝑉0 = = 0 𝑉 𝑉
2 V1 − V2 ( + ∆𝑉) − ( − ∆𝑉)
2 2
𝑉0 = = = +∆𝑉
2 2

Case-2: For FM signal frequency greater than carrier frequency ( f > fc )

When f > fc , the loop current in secondary winding of the input transformer, Is, will be leading
the secondary voltage, Vs, by an angle. Hence the vector sum of Vs1 and Vp i.e V1 is greater
than the vector sum of Vs2 and Vp i.e V2. So the output is a positive change from the output
voltage corresponding to the carrier frequency as shown in the vector diagram.
Case-3: For FM signal frequency less than carrier frequency ( f < fc )
When f < fc , the loop current in secondary winding of the input transformer, Is, will be lagging
the secondary voltage, Vs, by an angle. Hence the vector sum of Vs1 and Vp i.e V1 is less than
the vector sum of Vs2 and Vp i.e V2. So the output is a negative change from the output voltage
corresponding to the carrier frequency as shown in the vector diagram.

Case-3
𝑉 𝑉
𝑉1 < 𝑉2 , 𝑉1 = − ∆𝑉 And 𝑉2 = + ∆𝑉
2 2
𝑇ℎ𝑒𝑟𝑒𝑓𝑜𝑟𝑒
𝑉 𝑉
𝑉 𝑉 V1 − V2 (2 −∆𝑉)−( 2 +∆𝑉)
𝑉𝐶0 = 𝑉1 + 𝑉2 = 2 − ∆𝑉 + 2 + ∆𝑉 = V and 𝑉0 = 2
=
2
= −∆𝑉
Comparison between AM and FM:
S.NO Amplitude Modulation Frequency Modulation
1. Bandwidth is very small which is one o It requires much wider channel (7 to15 times)
the biggest advantage as compared to AM.
2. The amplitude of AM signal varies The amplitude of FM signal is constant and
depending on modulation index. independent of depth of the modulation.
3. Area of reception is large The area of reception is small since it is
limited to line of sight.
4. Transmitters are relatively simple & cheap. Transmitters are complex and hence
expensive.
5. The average power in modulated wave is The average power in frequency modulated
greater than carrier power. This added power wave is same as in unmodulated wave.
is provided by modulating source.
6. More susceptible to noise interference and Noise can be easily minimized. Amplitude
has low signal to noise ratio, it is more variations can be eliminated by using limiter.
difficult to eliminate effects of noise.
7. It is not possible to operate without It is possible to operate several independent
interference. transmitters on same frequency.
8. The maximum value of modulation index No restriction is placed on modulation index.
=1, otherwise over-modulation would result
in distortions.