EE3104C Introduction to RF and

Microwave Systems & Circuits

L4. Fundamentals of Amplifiers

Lecturer: Prof Guo Yongxin

Office location:E4-07-15
Office phone: 65162086
Email: [email protected]

1
 Outlines

• Introduction
• Dynamic range of a realistic amplifier
• White noise source
• Equivalent noise temperature
• Noise figure
• Nonlinear distortion

2
1. Introduction
GPA (General purpose Amp) / buffer amp

Amplifier
Specifications

3
Amplifier Specifications

4
Amplifier Specifications

5
Amplifier Specifications

6
 Design Sample 1
Low Noise Amplifier (HMC287)

- RF_IN and RF_OUT are matched to 50 Ohm hence can be connected directly to
the SMA port.
- A 3-way header is included to adjust the gain of the LNA to the maximum or
7
minimum value.
 Design Sample 2
General Purpose Amplifier (MNA-7)

- The input and output ports are matched to 50 Ohm

thus no matching circuits are required.
8
 A realistic amplifier - Dynamic range
• All realistic devices are nonlinear at very low power levels and at
high power levels.
• A dynamic range is defined as a range of signal levels where the
component is linear (meaning the output is directly proportional to
the input) and deterministic (meaning the output is predictable from
the input).

9
 Noise
• Even when the input power is zero, a certain nonzero noise power,
which is generated by the amplifier, is shown at the output.
• At very low input power levels, the output power will be dominated
by the noise of the amplifier. This level is often called the noise floor
of the component or system.
• Noise power is a result of random processes such as the flow of
charges or holes in an electron tube or solid-state device, propagation
through the ionosphere or other ionized gas, or, most basic of all, the
thermal vibrations in any component at a temperature above absolute
zero.

10
White noise Source
• A white noise source has a power
spectral density that is constant
with frequency.
N 0  KTB
where
k = 1.380 × 10−23 J/K is Boltzmann’s constant.
T = the temperature in degrees kelvin (K).
B = the bandwidth of the system in Hz.
 As B → 0, Pn → 0. This means that systems with smaller
bandwidths collect less noise power.
 As T → 0, Pn → 0. This means that cooler devices and
components generate less noise power.
 As B →∞, Pn →∞. This is the so-called ultraviolet catastrophe,
which does not occur in reality.
11
 Noise Power
Noise power is assumed to result from a matched resistor at To=290 K,
i.e., Ni=kToB. (Boltzmann’s constant: k=1.38*10-23 J/K)

Measurement Resolution BW (RBW)

RBW=8MHz RBW=30kHz 12
Equivalent noise temperature
• Assume an arbitrary white noise source delivers a noise power No
to a load resistor R. This noise source can be replaced by a noisy
resistor of value R at temperature Te, where Te is an equivalent
temperature so that the same noise power is delivered to the load.

N0
Te 
kB

13
Equivalent noise temperature
• A noisy amplifier with a bandwidth B
and gain G is matched to noiseless
source and load resistors.
 If the source resistor is at a
(hypothetical) temperature of Ts = 0
K, then the input power to the
amplifier will be Ni = 0, and the
output noise power No will be due
only to the noise generated by the
amplifier itself.
 We can obtain the same load noise
power by driving an ideal noiseless
amplifier with a resistor at the
temperature Te, to make No=GkTeB.
 Te is the equivalent noise temperature
of the amplifier. 14
Noise figure
• The noise figure, F, of the component, which is a measure of the
degradation in the signal-to-noise ratio between the input and output
of the component.
S /N
F  i i 1
So / N o
where Si , Ni are the input signal and noise powers, and So, No are the
output signal and noise powers.
 The input noise power is assumed to be the noise power resulting
from a matched resistor at T0 = 290 K; that is, Ni = kT0B.

15
Noise figure
Input noise power is Ni=kT0B.
Output signal power is So=GSi.
Output noise power is
N o  kGB(T0  Te )

Si / N i Si kGB (T0  Te ) Te
F   1  1  T0  Te  FT0
So / N o kT0 B GSi T0
N o  FkGBT0

• Te can be obtained, Te  ( F  1)T0

• The noise figure in dB, F = 10 log(1+Te/T0) dB ≥ 0.
 If the network were noiseless, Te would be zero, giving F = 1,
or 0 dB.
16
 Example
• Power gain of an amplifier is 20 dB in the frequency band of 10 GHz
to 12 GHz. If its noise figure is 3.5 dB, find the output noise power in
dBm.
Solution:
Output noise power No,
N o  kGB(T0  Te )
  N o  FGkBT0
Te  ( F  1)T0 
F  3.5dB  100.35  2.2387
G  20dB  10 2  100
  
N o  2.2387 100  1.38 10  23  2 109  290
 1.7919 10 9 W  1.7919 10 6 mW
 
 10 log 1.7919 10 6 dBm  57.47 dBm
17
 Noise figure of a cascaded system
• Consider the cascade of two components, having gains G1, G2, noise
figures F1, F2, and equivalent noise temperatures Te1, Te2,.

N1  G1kT0 B  G1kTe1B
N o  G2 N1  G2 kTe 2 B  G2 G1kB(T0  Te1 )  G2 kTe 2 B
 G1G2 kB T0  Te1  Te 2 / G1   G1G2 kB T0  Tcas 
• The noise temperature of the cascaded system, Tcas
T e2
Tcas  Te1 
G1
F2  1
From Te1  ( F1  1)T0 and Te 2  ( F2  1)T0 , we get Fcas  F1 
G1 18
 Noise figure of a cascaded system
• For a general system with an arbitrary number of stages
T e 2 T e3
Tcas  Te1   
G1 G1G2
F2  1 F3  1
Fcas  F1   
G1 G1G2
• The noise characteristics of a cascaded system are dominated by the
characteristics of the first stage since the effect of the second stage
is reduced by the gain of the first stage (assuming G1 > 1).
• For the best overall system noise performance, the first stage should
have a low noise figure and at least moderate gain (Low Noise
Amplifier!).

19
 Example
Two amplifiers, each with 20-dB gain, are connected in cascade as shown
below. The noise figure of amplifier A1 is 3 dB while that of A2 is 5 dB.
Calculate the overall gain and noise figure for this arrangement. If the
order of two amplifier is changed in the system, find its resulting noise
figure.

Solution: The noise factors and gains of two amplifiers are:

F1  3 dB  100.3  2
F2  5 dB  100.5  3.1623
G1  G2  20 dB  102  100
20
 Example (ctd)
Therefore, the overall gain and noise figure of the cascaded system is
found as follows:
P3 P3 P2
G     100 100  40 dB
P1 P2 P1
F2  1 3.1623  1
F  F1   2  2.021623  3.057 dB
G1 100

If the order of two amplifier is changed then the overall gain will stay the
same. However, the noise figure of new arrangement will change as
follows.

F1  1 2 1
F  F2   3.1623   3.1723  5.013 dB
G1 100

21
 Example
• Compute the overall noise figure of this subsystem. If the input noise
power from a feeding antenna is Ni = kTAB, where TA = 150 K, find the
output noise power in dBm. If we require a minimum signal-to-noise
ratio (SNR) of 20 dB at the output of the receiver, what is the
minimum signal power that should be applied at the receiver input?
Assume the system is at temperature T0, with a characteristic
impedance of 50 ꭥ, and an IF bandwidth of 10 MHz.

22
 Example
Solution:
Ga  10dB  10 G f  1dB  0.79 Gm  3dB  0.5
Fa  2dB  1.58 F f  1dB  1.26 Fm  4dB  2.51
The system noise figure:
F f  1 Fm  1 1.26  1 2.51  1
Fcas  Fa    1.58    1.80  2.55dB
Ga Ga G f 10 10  0.79
The system equivalent noise temperature:
Tcas  Fcas  1T0  1.80  1 290  232 K
The output noise power:
N o  k (TA  Tcas ) BG  k (TA  Tcas ) BGa G f Gm
   
 1.38 10-23 150  232  10 106 10 0.79 0.5
 2.08 10 13 W  96.8 dBm
23
 Example
For an output SNR of 20 dB,
S o  N 0  20dB  100 N o  2.08 10 11 W
So So 2.08 10 11
Si     5.27 10 12 W  82.2 dBm
G G a G f Gm 3.95

24
 Nonlinearities in Amplifiers
• The output response of a nonlinear circuit can be modeled as a
Taylor series expansion up to its third order term in terms of the
input signal voltage

𝑣𝑜 = 𝑐0 + 𝑐1 𝑣𝑖 + 𝑐2 𝑣𝑖2 + 𝑐3 𝑣𝑖3

• Impact of nonlinearities with sine wave input

- Causes harmonic distortion (i.e., creation of harmonics)
vi  A cos(t )
• Impact of nonlinearities with several sine wave inputs
- Causes harmonic distortion for each input AND
inter-modulation products
vi  A(cos(1t )  cos(2t ))
25
 Harmonic Distortion
𝑣𝑜 = 𝑐0 + 𝑐1 𝑣𝑖 + 𝑐2 𝑣𝑖2 + 𝑐3 𝑣𝑖3
where, vi  A cos(t )
• Substitute x(t) into polynomial expression
𝑣𝑜 = 𝑐0 + 𝑐1 𝐴 cos 𝜔 𝑡 + 𝑐2 𝐴2 cos 2 𝜔 𝑡 + 𝑐𝐴3 cos 3 𝜔 𝑡
(1 + cos 2 𝜔𝑡) 3 cos 𝜔 𝑡 + cos( 3𝜔𝑡)
= 𝑐0 + 𝑐1 𝐴 cos 𝜔 𝑡 + 𝑐2 𝐴2 + 𝑐3 𝐴3
2 4
c2 𝐴2 3𝑐3 𝐴3 𝑐2 𝐴2 𝑐3 𝐴3
= (𝑐0 + ) + (𝑐1 𝐴 + ) cos 𝜔 𝑡 + cos 2 𝜔𝑡 + cos 3 𝜔𝑡
2 4 2 4

Fundamental Harmonics
•Notice that each harmonic term, cos(nωt), has an amplitude that
grows in proportion to An
- Very small for small A, very large for large A
26
 1 dB Compression Point

3
𝐺
At fundamental frequency ω0, 𝑣 = 𝑐1 + 𝑐3 𝐴2
4
 c3 typically has the opposite sign of a1, so
that the output of the amplifier tends to be
reduced from the expected linear
dependence for large values of A.
 This effect is called gain compression, or
saturation.
 1 dB compression point is defined as the power level for which the output power
has decreased by 1 dB from the ideal linear characteristic.
 This power level is usually denoted by P1dB, and can be stated in terms of either
input power (IP1dB) or output power (OP1dB).
OP1dB (dB)  IP1dB (dB)  G (dB )  1 27
Harmonic Products with An Input of two Sine Waves

y (t )  c0  c1 x(t )  c2 x(t ) 2  c3 x(t ) 3

where, x(t )  A(cos(1t )  cos(2t ))
• A two-tone input voltage with two closely spaced frequencies,
• DC and fundamental components

 2
 9 
c0  c2 A   (c1 A  c3 A )(cos 1t  cos 2t ) 
3

 4 
•Second and third harmonic terms
 c2 A2   c3 A3 
 (cos 21t  cos 22t )    (cos 31t  cos 32t ) 
 2   4 
•Similar result as having an input with one sine wave
-But, we haven’t yet considered cross terms! 28
Harmonic Products with An Input of two Sine Waves

y(t )  c0  c1 x(t )  c2 x(t )  c3 (t )

2 3

where, x(t )  A(cos(1t )  cos(2t ))

• Second-order inter-modulation(IM2) products

c2 A cos(1  2 )t  cos(2  1 )t 
2

•Third-order inter-modulation(IM3) products

3
c3 A3 (cos(21  2 )t  cos(21  2 )t
4
 cos(22  1 )t  cos(22  1 )t )
-These are the troublesome ones for narrowband systems
29
 Two-tone input voltage
 The output spectrum consists of harmonics of the form mω1+nω2,
with m, n=0, 1, 2, 3,…. These combinations of the two input
frequencies are called intermodulation products, and the order of a
given product is defined as |m|+|n|.
 There are 4 third-order intermodulation products: 2ω1 + ω2, 2ω2 +
ω1, 2ω1 − ω2, and 2ω2 − ω1.
 The first 2 of these will again be located far from ω1 or ω2, and
will typically be outside the passband of the component.
 The two difference terms produce products located near the
original input signals at ω1 and ω2, and so cannot be easily
filtered from the passband of an amplifier.→ third-order
intermodulation distortion

30
 Third-order intercept point
• The input voltage A increases, the voltage associated with the third-
order products increases as A3 . Since power is proportional to the
square of voltage, the output power of third-order products must
increase as the cube of the input power.
• The output power of the
first-order, or linear,
product is proportional to
the input power, and so the
line describing this
response has a slope of
unity.
• The line describing the
response of the third-order
products has a slope of 3.
31
 Third-order intercept point
• Both the linear and third-order responses will exhibit compression at
high input powers, so we show the extension of their idealized
responses with dotted lines.
• Since these two lines have different slopes, they will intersect..
 This hypothetical
intersection point where
the first-order and third-
order powers would be
equal is called the third-
order intercept point,
denoted as IP3.
 It may be specified as
either an input power level
(IIP3), or an output power
level (OIP3).
OIP3 = G (IIP3). 32
Corruption of Narrowband Signals by Interferers

-LNA nonlinearity causes the creation of harmonic and inter-

modulation products
-Must remove interference and its products to retrieve desired signal

33
Use Filtering to Remove Undesired Interference

• Ineffective for IM3 term that falls in the desired signal frequency band
34
 Characterization of Inter-modulation
• Magnitude of third order products is set by c3 and input signal
amplitude (for small A)
3
c3 A3 (cos(21  2 )t  cos(21  2 )t
4
 cos(22  1 )t  cos(22  1 )t )
•Magnitude of first order term is set by c1and A (for small A)
9
(c1 A  c3 A3 )(cos 1t  cos 2t )  c1 A(cos 1t  cos 2t )
4
•Relative impact of inter-modulation products can be calculated
once we know A and the ratio of c3 to c1
- Problem: it’s often hard to extract the polynomial coefficients
through direct DC measurements
- Need an indirect way to measure the ratio of c3 to c1 35
 Relationship between IIP3, c1and c3
• Intersection point
3
c1 A  c3 A3
4
• Solve for A (gives Aiip3)

4 c1
A  2

3 c3

•Note that A corresponds to the peak value of the two cosine waves
coming into the amplifier

36
 Two Tone Test
• Input the sum of two equal amplitude sine waves into the amplifier
(assume Zin of amplifier = Rs of source)

• On a spectrum analyser, measure first order and third order terms as A is

varied (A must remain small)
- First order term will increase linearly
-Third order IM term will increase as the cube of A 37
 Third Order Intercept Point (IIP3)
• Plot the results of the two-tone test over a range of A (where A remains
small) on a log scale (i.e., dB)
-Extrapolate the results to find the intersection of the first and third order
terms

- IIP3 defined as the input power at which the extrapolated lines intersect
(higher value is better)
• Note that IIP3 is a small signal parameter based on extrapolation, in 38
contrast to the 1-dB compression point
 Measuring IMD Performance

• Set the amplitude of generators at f1 and f2 to be equal.

• Pin must be more than 10 dB below P1dB .
-Start at a very low input power using the variable attenuator,
then increase power in steps until you begin to see the IMD
output on the spectrum analyzer.
• Use narrow RBW on SA to reduce noise floor
-This will allow visibility of the IMD signal at lower power levels.
• Plot the IMD power vs. input power and verify that the slope is
close to 3. Then, you can calculate the IP3 as described previously.
39
Conclusion
 A realistic amplifier
Dynamic range, noise, and gain compression
 White noise source
 Equivalent noise temperature
 Noise figure
A single stage system
A cascaded system
 Nonlinear distortion
A single frequency input, two-tone input voltage, third order
intersect point, and two tone test.

40