0% found this document useful (0 votes)
110 views54 pages

Day7 PA Share PDF

The document discusses amplifier stability analysis. It explains how a two-port network model can represent an amplifier inserted into a transmission line between a source and load. Stability depends on the input and output reflection coefficients, which are functions of the scattering parameters and source/load impedances. Unconditional stability requires both reflection coefficients to be less than 1 for all passive source and load impedances. Stability circles in the load/source impedance planes define boundaries between stable and potentially unstable regions. Resistive loading can improve stability by making the total input and output loop resistances positive.

Uploaded by

MM GOPAL
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
110 views54 pages

Day7 PA Share PDF

The document discusses amplifier stability analysis. It explains how a two-port network model can represent an amplifier inserted into a transmission line between a source and load. Stability depends on the input and output reflection coefficients, which are functions of the scattering parameters and source/load impedances. Unconditional stability requires both reflection coefficients to be less than 1 for all passive source and load impedances. Stability circles in the load/source impedance planes define boundaries between stable and potentially unstable regions. Resistive loading can improve stability by making the total input and output loop resistances positive.

Uploaded by

MM GOPAL
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 54

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.

com

www.vlsi-expert.com

AMPLIFIER
www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Theory

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com
Transmission Line Connected Two-port Network inserted into the line
between source and load b r 1 = S 11 a i1 + S 12 a i2
www.vlsi-expert.com
b r 2 = S 21 a i1 + S 22 a i2

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Theory

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com
ZS Zin Zout ZL

www.vlsi-expert.com
 Determine ZS and ZL  Use Lumped Elements

 Use Distributed Elements


www.vlsi-expert.com
Match ZS to Z0 www.vlsi-expert.com
 Use Combination of Lumped and
 Matchwww.vlsi-expert.com
ZL to Z0
Distributed Elements

www.vlsiexpert.com | www.technoready.com
Amplifier Circuit

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com
1 − Γ𝑠𝑠 2 1 − Γ𝐿𝐿 2
𝐺𝐺𝑡𝑡 = 𝑆𝑆21
1 − Γ𝑖𝑖𝑖𝑖 Γ𝑠𝑠 2 1 − 𝑆𝑆22 Γ𝐿𝐿 2
www.vlsi-expert.com OR www.vlsi-expert.com

www.vlsi-expert.com
− Γ𝑠𝑠 2
1 1 − Γ𝐿𝐿 2
𝐺𝐺𝑡𝑡 = 𝑆𝑆21
1 − 𝑆𝑆11 Γ𝑠𝑠 2 1 − Γ𝑜𝑜𝑜𝑜𝑜𝑜 Γ𝐿𝐿 2

www.vlsiexpert.com | www.technoready.com
Stability Considerations

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

𝑆𝑆12 𝑆𝑆21 Γ𝐿𝐿 𝑆𝑆12 𝑆𝑆21 Γ𝑠𝑠


Γ𝑖𝑖𝑖𝑖 = 𝑆𝑆11 + Γ𝑜𝑜𝑜𝑜𝑜𝑜 = 𝑆𝑆22 +
www.vlsi-expert.com
1 − 𝑆𝑆22 Γ𝐿𝐿 1 − 𝑆𝑆22 Γ𝑠𝑠
In the above circuit, oscillation is possible if either the input
www.vlsi-expert.com www.vlsi-expert.com
or output port impedance has a negative real value
www.vlsi-expert.com
Γin > 1 or Γout > 1 Stability of the Amplifier Depends on
ΓS and ΓL
www.vlsiexpert.com | www.technoready.com
Stability Considerations

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

Unconditional Stability: The network is unconditionally stable if


www.vlsi-expert.com
Γ
𝑖𝑖𝑖𝑖 < 1 𝑎𝑎𝑎𝑎𝑎𝑎 𝑜𝑜𝑜𝑜𝑜𝑜 < 1
Γ

for all passive source and load impedances. (ie Γ𝑠𝑠 < 1 𝑎𝑎𝑎𝑎𝑎𝑎 Γ𝐿𝐿 < 1)
www.vlsi-expert.com www.vlsi-expert.com
Conditional Stability: The network is conditionally stable if
Γ
𝑖𝑖𝑖𝑖 < 1 𝑎𝑎𝑎𝑎𝑎𝑎 Γ𝑜𝑜𝑜𝑜𝑜𝑜 < 1
www.vlsi-expert.com
for certain range of passive source and load impedances. This case is also
referred as potential unstable. (ie Γ𝑠𝑠 < 1 𝑎𝑎𝑎𝑎𝑎𝑎 Γ𝐿𝐿 < 1)
www.vlsiexpert.com | www.technoready.com
Input and Output Stability Circles
𝑆𝑆12 𝑆𝑆21 Γ𝐿𝐿 𝑆𝑆12 𝑆𝑆21 Γ𝑠𝑠
Γ𝑖𝑖𝑖𝑖 = 𝑆𝑆11 + <1 Γ𝑜𝑜𝑜𝑜𝑜𝑜 = 𝑆𝑆22 + <1
1 − 𝑆𝑆22 Γ𝐿𝐿 www.vlsi-expert.com 1 − 𝑆𝑆 Γ
www.vlsi-expert.com
22 𝑠𝑠
www.vlsi-expert.com

Stability Circles: Defined as the loci in the ΓL(or Γs ) Plane for which Γ𝑖𝑖𝑖𝑖 = 1 𝑜𝑜𝑜𝑜 Γ𝑜𝑜𝑜𝑜𝑜𝑜 = 1 .
These circles define the boundaries between stable and potentially unstable regions of Γ𝑠𝑠 𝑎𝑎𝑎𝑎𝑎𝑎 Γ𝐿𝐿
www.vlsi-expert.com

∆𝑆𝑆11 ∗ ∗
𝑆𝑆22 −www.vlsi-expert.com 𝑆𝑆12 𝑆𝑆21
OUTPUT STABILITY CIRCLE 𝐶𝐶𝐿𝐿 = 𝑅𝑅𝐿𝐿 =
2
𝑆𝑆22 − ∆ 2 𝑆𝑆22 2 − ∆ 2

www.vlsi-expert.com www.vlsi-expert.com
∗ ∗
𝑆𝑆11 − ∆𝑆𝑆22 𝑆𝑆12 𝑆𝑆21
INPUT STABILITY CIRCLE
www.vlsi-expert.com 𝐶𝐶𝑠𝑠 = 𝑅𝑅𝐿𝐿 =
𝑆𝑆11 2 − ∆ 2 𝑆𝑆11 2 − ∆ 2

www.vlsiexpert.com | www.technoready.com
∆= 𝑆𝑆11 𝑆𝑆22 − 𝑆𝑆12 𝑆𝑆21
Region of Stability

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

𝑆𝑆11 < 1 𝑆𝑆11 > 1


www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Unconditional Stability Conditions

1 + 𝑆𝑆11 𝑆𝑆22 − 𝑆𝑆12 𝑆𝑆21 2 − 𝑆𝑆11 2 2


− 𝑆𝑆22www.vlsi-expert.com www.vlsi-expert.com
𝐾𝐾 = www.vlsi-expert.com ∆= 𝑆𝑆11 𝑆𝑆22 − 𝑆𝑆12 𝑆𝑆21
2 𝑆𝑆12 𝑆𝑆21
www.vlsi-expert.com
For unconditional Stability, the necessary and sufficient conditions are that

∆ < 1 𝑎𝑎𝑎𝑎𝑎𝑎 𝐾𝐾 < 1


www.vlsi-expert.com

Recent Combined Formula to determine stability


www.vlsi-expert.com www.vlsi-expert.com
1− 𝑆𝑆11 2
www.vlsi-expert.com 𝜇𝜇 = >1

𝑆𝑆22 − 𝑆𝑆11 ∆ + 𝑆𝑆12 𝑆𝑆21

www.vlsiexpert.com | www.technoready.com
Stability Improvement Using resistive loading

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

To make device unconditionally stable, total input and output loop


www.vlsi-expert.com
resistances should be positive
Re (Zs + Zin ) > 0; Re(ZL + Zout ) > 0
www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com
To achieve this, we use resistive loadings at the input and output of the
amplifier
www.vlsiexpert.com | www.technoready.com
Stability Improvement Using resistive loading

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Single Stage Amplifier Design
1. Given the device S-parameters, Draw Stability Circles on the Smith Chart.
2. www.vlsi-expert.com
Since
www.vlsi-expert.com www.vlsi-expert.com
Gd is fixed, the overall amplifier gain is controlled by the gains Gs and GL.

3. Maximum gain will be realized when the matching section provide a conjugate
www.vlsi-expert.com
match at the input and output.
Γ𝑖𝑖𝑖𝑖 =Γ𝑠𝑠 ∗ 𝑎𝑎𝑎𝑎𝑎𝑎 Γ𝑜𝑜𝑜𝑜𝑜𝑜 = Γ𝐿𝐿 ∗
www.vlsi-expert.com
4. Determine Γs and ΓL
5. Design Suitable matching networks at the I/O
www.vlsi-expert.com www.vlsi-expert.com
1 1 − Γ𝐿𝐿 2
𝐺𝐺𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡
www.vlsi-expert.com = 𝑆𝑆21
1 − Γ𝑠𝑠 2 1 − S22 Γ𝐿𝐿 2

www.vlsiexpert.com | www.technoready.com
Design for Specific Gain-Constant Gain Circles

OBJECTIVE: To Improve Bandwidth or to obtain a specific value of amplifier gain, it is


sometimes preferred to design amplifier for less
www.vlsi-expert.com
than the maximum
www.vlsi-expert.com obtainable gain.
www.vlsi-expert.com
APPROACH:
www.vlsi-expert.com
 Redesign the input and output matching sections to have less than maximum
gains
www.vlsi-expert.com
 Introduce mismatches intentionally to reduce overall gain

 Design procedure is simplified by plotting constant gain circles on the smith


www.vlsi-expert.com www.vlsi-expert.com
chart
www.vlsi-expert.com
 Gain Circles: Loci of Γs and ΓL that give fixed value of gain GS or GL

www.vlsiexpert.com | www.technoready.com
Constant Gain Circle

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Constant Gain Circle

OUTPUTwww.vlsi-expert.com
CONSTANT 𝑔𝑔𝐿𝐿 𝑆𝑆22 ∗
www.vlsi-expert.com www.vlsi-expert.com
1 − 𝑔𝑔𝐿𝐿 1 − 𝑆𝑆22 2
GAIN CIRCLE 𝐶𝐶𝑔𝑔𝑔𝑔 = 𝑅𝑅𝑔𝑔𝑔𝑔 =
1− 1 − 𝑔𝑔𝐿𝐿 𝑆𝑆22 2 1 − 1 − 𝑔𝑔𝐿𝐿 𝑆𝑆22 2
www.vlsi-expert.com

www.vlsi-expert.com
INPUT CONSTANT
𝑔𝑔𝑠𝑠 𝑆𝑆11 ∗ 1 − 𝑔𝑔𝑠𝑠 1 − 𝑆𝑆11 2
𝐶𝐶𝑔𝑔𝑔𝑔 = 𝑅𝑅𝑔𝑔𝑔𝑔 =
GAIN CIRCLE 1 − 1 − 𝑔𝑔𝐿𝐿 𝑆𝑆11 2 1 − 1 − 𝑔𝑔𝑠𝑠 𝑆𝑆11 2
www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
EXAMPLE
Q - Design an amplifier to have gain of 11 dB at 4.0 GHz. Plot constant gain circles for
GS=2 dB and 3 dB and GL=0dB and 1dB. Design matching networks using single stub and
piece of transmission www.vlsi-expert.com
line. Calculate the gain
www.vlsi-expert.com www.vlsi-expert.com
from 3 to 5 GHz. The GaAs FET has the
following S-parameters (Z0= 50 ohm)

www.vlsi-expert.com
F(GHz) S11 S21 S12 S22

3.0 0.80∠ −900 2.8∠1000 0 0.66∠ − 500


www.vlsi-expert.com

4.0 0.75∠ − 1200 2.5∠800 0 0.60∠ − 700


www.vlsi-expert.com www.vlsi-expert.com
5.0 0.71∠ − 140 0 2.3∠600 0 0.58∠ − 850
www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
EXAMPLE
Q - Design an amplifier to have gain of 11 dB at 4.0 GHz. Plot constant gain circles for
GS=2 dB and 3 dB and GL=0dB and 1dB. Design matching networks using single stub and
piece of transmission www.vlsi-expert.com
line. Calculate the gain
www.vlsi-expert.com www.vlsi-expert.com
from 3 to 5 GHz. The GaAs FET has the
following S-parameters (Z0= 50 ohm)

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
EXAMPLE

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Power Gain Concepts

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com
P in Power input to the matching network or to the transistor.

𝑃𝑃𝐴𝐴 www.vlsi-expert.com
- Power delivered to the load or to the outputwww.vlsi-expert.com
matching network

case of 𝑷𝑷𝒊𝒊𝒊𝒊 𝒘𝒘𝒘𝒘𝒘𝒘𝒘𝒘 Γ𝑖𝑖𝑖𝑖 =Γ𝑠𝑠 ∗


www.vlsi-expert.com
𝑃𝑃𝑎𝑎𝑎𝑎𝑎𝑎 - Special

𝑃𝑃𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 - Power available from transistor when Γ𝑜𝑜𝑜𝑜𝑜𝑜 =Γ𝐿𝐿 ∗


www.vlsiexpert.com | www.technoready.com
Power Gain Concepts

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com
𝑷𝑷𝑨𝑨
Transducer Power Gain: 𝑮𝑮𝒕𝒕 =
𝑷𝑷𝒂𝒂𝒂𝒂𝒂𝒂
www.vlsi-expert.com www.vlsi-expert.com
𝑷𝑷𝑨𝑨
Operating Power Gain: 𝑮𝑮𝑷𝑷 =
𝑷𝑷𝒊𝒊𝒊𝒊
www.vlsi-expert.com
𝑷𝑷𝒂𝒂𝒂𝒂𝒂𝒂𝒂𝒂
Available Power Gain: 𝑮𝑮𝒂𝒂 =
𝑷𝑷𝒂𝒂𝒂𝒂𝒂𝒂

www.vlsiexpert.com | www.technoready.com
DESIGN FORMULAE

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
DESIGN FORMULAE

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
DESIGN FORMULAE

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Trade off Example
The S-parameters of the GaAs FET at 12 GHz are
S11 = 0.6∠360, S12 = 0.14∠−850 ,S21 = 2.3∠−800 ,S22 = 0.15∠450
www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com
K = 1.17 and Δ = 0.368
Transistor is unconditionally stable
www.vlsi-expert.com
Γ𝑀𝑀𝑀𝑀 = 0.714⌊− 40.450
Γ𝑀𝑀𝑀𝑀 = 0.387⌊129.360
VSWR at IN and OUT ports=1
www.vlsi-expert.com
𝑆𝑆21
𝐺𝐺𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = 𝐾𝐾 − 𝐾𝐾 2 − 1 = 9.24 9.66 𝑑𝑑𝑑𝑑
𝑆𝑆12
Maximum Value of Associated Gain
www.vlsi-expert.com www.vlsi-expert.com
𝐺𝐺 𝐴𝐴 = 8.66𝑑𝑑𝑑𝑑 Design Goal for Gain
www.vlsi-expert.com
Ca = 0.602∠ − 40.45 ; 0

ra = 0.3
www.vlsiexpert.com | www.technoready.com
Trade off Example
The S-parameters of the GaAs FET at 12 GHz are
S11 = 0.6∠360, S12 = 0.14∠−850 ,S21 = 2.3∠−800 ,S22 = 0.15∠450
www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com
K = 1.17 and Δ = 0.368
Transistor is unconditionally stable
www.vlsi-expert.com
Γ𝑀𝑀𝑀𝑀 = 0.714⌊− 40.450
Γ𝑀𝑀𝑀𝑀 = 0.387⌊129.360
VSWR at IN and OUT ports=1
www.vlsi-expert.com
𝑆𝑆21
𝐺𝐺𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = 𝐾𝐾 − 𝐾𝐾 2 − 1 = 9.24 9.66 𝑑𝑑𝑑𝑑
𝑆𝑆12
Maximum Value of Associated Gain
www.vlsi-expert.com www.vlsi-expert.com
𝐺𝐺 𝐴𝐴 = 8.66𝑑𝑑𝑑𝑑 Design Goal for Gain
www.vlsi-expert.com
Ca = 0.602∠ − 40.45 ; 0

ra = 0.3
www.vlsiexpert.com | www.technoready.com
Trade off Example
The S-parameters of the GaAs FET at 12 GHz are
S11 = 0.6∠360, S12 = 0.14∠−850 ,S21 = 2.3∠−800 ,S22 = 0.15∠450
www.vlsi-expert.com www.vlsi-expert.com For
www.vlsi-expert.com
K = 1.17 and Δ = 0.368
Transistor is unconditionally stable
Γ𝑀𝑀𝑀𝑀 = 0.714⌊− 40.450
www.vlsi-expert.com VSWRout =1.5
Γ𝑀𝑀𝑀𝑀 = 0.387⌊129.360
VSWR at IN and OUT ports=1
www.vlsi-expert.com
𝑆𝑆21
𝐺𝐺𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴 = 𝐾𝐾 − 𝐾𝐾 2 − 1 = 9.24 9.66 𝑑𝑑𝑑𝑑
𝑆𝑆12
Maximum Value of Associated Gain
www.vlsi-expert.com www.vlsi-expert.com
𝐺𝐺 𝐴𝐴 = 8.66𝑑𝑑𝑑𝑑 Design Goal for Gain
www.vlsi-expert.com
Ca = 0.602∠ − 40.45 ; 0

ra = 0.3
www.vlsiexpert.com | www.technoready.com
www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
General Non-linear Device or Network

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

• All realistic components have a small loss- thermal noise generated by a lossy component
- Devices are non-linear at very low power levels due to noise effects
www.vlsi-expert.com
• At high power levels, we have gain compression or generation of spurious frequency components
- Devices are non-linear at high power levels as well
www.vlsi-expert.com www.vlsi-expert.com
These effect set a minimum and maximum realistic power range
www.vlsi-expert.com
(Dynamic Range)
over which a component will operate as desired
www.vlsiexpert.com | www.technoready.com
Output Response of a Non-linear Circuit
Response is approximated by Taylor series:

v = a + a v + a v 2 + a v 3 + .......
www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com
0 0 1 i 2 i 3 i

www.vlsi-expert.com
Where Taylor coefficients are given by:

a0 = v 0 (0) (DC Output) Rectifier


www.vlsi-expert.com
dv0
a1 = (Linear Output) Attenuator a1<1
dv i vi =0 Amplifier a1>1
d2 vwww.vlsi-expert.com www.vlsi-expert.com
a = 0
(Squared Output)
2 Mixer etc.
dvi v =0 www.vlsi-expert.com
2
i

www.vlsiexpert.com | www.technoready.com
Gain Compression
Single Frequency Sinusoid Applied to the Non-linear Circuit vi = V0 cosω0t

a V cos ω t + a V 2 cos2 ω t + a V 3 coswww.vlsi-expert.com


v0 = a +www.vlsi-expert.com 3
ω t +....... www.vlsi-expert.com

1   3 3  1
v 0 =  a 0 + a 2 V 02  +www.vlsi-expert.com
a V
 1 0 + a V
3 0 
cos ω 0
t + a 2 V02 cos 2 ω0 t
 2   4  2
1
+ a 3 V 03 cos 3 ω 0 t + .......
4 www.vlsi-expert.com

3 2
𝑎𝑎 𝑉𝑉 + 𝑎𝑎
𝑣𝑣0 www.vlsi-expert.com
1 𝑜𝑜 𝑉𝑉
3 𝑜𝑜
𝐺𝐺𝑣𝑣 = = 4 www.vlsi-expert.com
𝑣𝑣𝑖𝑖 𝑉𝑉𝑜𝑜
www.vlsi-expert.com
3
= 𝑎𝑎1 + 𝑎𝑎3 𝑉𝑉 2 𝑜𝑜
4 P1(dB) – in terms of output power for amplifiers and in
terms of input power for mixer
www.vlsiexpert.com | www.technoready.com
Inter Modulation Distortion
Single Frequency Sinusoid Applied to the Non-linear Circuit vi = V0 cosω0t

a V cos ω t + a V 2 cos2 ω t + a V 3 coswww.vlsi-expert.com


v0 = a +www.vlsi-expert.com 3
ω t +....... www.vlsi-expert.com

1   3 3  1
v 0 =  a 0 + a 2 V 02  +www.vlsi-expert.com
a V
 1 0 + a V
3 0 
cos ω 0
t + a 2 V02 cos 2 ω0 t
 2   4  2
1
+ a 3 V 03 cos 3 ω 0 t + .......
4 www.vlsi-expert.com

The first term – Represents dc Voltage [useful in a rectifier application


www.vlsi-expert.com www.vlsi-expert.com
Second term – Primarily for amplifier application
www.vlsi-expert.com
Last two terms – useful for frequency multiplier circuits

www.vlsiexpert.com | www.technoready.com
Single - Tone Input
Single Frequency Sinusoid Applied to the Non-linear Circuit vi = V0 cosω0t

a V cos ω t + a V 2 cos2 ω t + a V 3 coswww.vlsi-expert.com


v0 = a +www.vlsi-expert.com 3
ω t +....... www.vlsi-expert.com

1   3 3  1
v 0 =  a 0 + a 2 V 02  +www.vlsi-expert.com
a V
 1 0 + a V
3 0 
cos ω 0
t + a 2 V02 cos 2 ω0 t
 2   4  2
1
+ a 3 V 03 cos 3 ω 0 t + .......
4 www.vlsi-expert.com

Presence of other frequency components lead to signal distortion if these


are in www.vlsi-expert.com
the pass band of the amplifier www.vlsi-expert.com

consists of harmonics of input frequency of the form nω0


www.vlsi-expert.com
 Output
 Usually these harmonics do not interfere with the desired signal at ω0

www.vlsiexpert.com | www.technoready.com
Two - Tone Input
Two tone Frequency Sinusoid Applied to the Non-linear Circuit : v i = V0 (cos ω1 t + cos ω2 t )

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com


1 1
a 2 V02 (1 + cos 2 ω1 t ) + a 2 V02 (1 + cos 2 ω 2 t )
Squared terms: 2 2
+ a 2 V02 cos (ω 1 − ω 2 ) t + a 2 V02 cos (ω 1 + ω 2 ) t
www.vlsi-expert.com

Four Inter-modulation products of second order


www.vlsi-expert.com
• These second order products are undesired in
an amplifier
• In a mixer, www.vlsi-expert.com
sum and difference products are www.vlsi-expert.com
desired outputs
www.vlsi-expert.com
𝜔𝜔1 is close to𝜔𝜔2 ; sec ond order products can be
filtered
www.vlsiexpert.com | www.technoready.com
Two - Tone Input
Two tone Frequency Sinusoid Applied to the Non-linear Circuit : v i = V0 (cos ω1 t + cos ω2 t )

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Two - Tone Input
Two tone Frequency Sinusoid Applied to the Non-linear Circuit : v i = V0 (cos ω1 t + cos ω2 t )

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

Input signal consisting of many frequencies of varying amplitude and phase will result
www.vlsi-expert.com
in distortion of output signal- 3rdorder inter-modulation distortion
www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Third Order Intercept Point

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com
Intersection point where first order and third order powers are
equal is called 3rd order Intercept point [pointP3]
www.vlsi-expert.com
P3 is referenced at the output for amplifiers and at the input for mixers :
Usually P3 is 12-15 dB greater thanP1
www.vlsiexpert.com | www.technoready.com
Linear and Spurious-free Dynamic Range
Dynamic Range: Operating range for which a component has desirable characteristics

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com


Linear Dynamic Range:
 For power amplifier, the power range may
www.vlsi-expert.com
be limited at low end by noise and at the
high end by the compression point

𝐷𝐷𝐷𝐷𝑙𝑙 = 𝑃𝑃1 − 𝑁𝑁𝑜𝑜


www.vlsi-expert.com
Spurious Free Dynamic Range:
 For low noise amplifier and mixer, the
operationwww.vlsi-expert.com
may be limited at low end by www.vlsi-expert.com
noise and at the maximum power level for
www.vlsi-expert.com
which inter-modulation distortion becomes
unacceptable.
2
𝐷𝐷𝐷𝐷𝑓𝑓 = 𝑃𝑃3 − 𝑁𝑁𝑜𝑜
3 www.vlsiexpert.com | www.technoready.com
Device utilization factor

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Efficiency Matrix of a PA

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Why Efficiency?

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Power Amplifier Design
Power Amplifier: An Amplifier that is designed to deliver maximum output power for a given
active device.
www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

Difference as compare to Linear RF Amplifier


www.vlsi-expert.com
design:
 In Linear RF Amplifier design both the input
and output are conjugate match in order to
achieve maximum available gain.
www.vlsi-expert.com

 In Power Amplifier (PA) design the input is


conjugatewww.vlsi-expert.com
match an the output is power www.vlsi-expert.com
match in order to achieve maximum output
power. www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Effect of Output Termination on PA performance

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Typical behaviour of PA with Conjugate Match
and Power Match
 Practically, conjugate match implementation is limited by: The current supplied by
the device and the voltage that the device can sustain across its terminals without
any damage.
www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

 From Fig. Linear output power range is


higher in powerwww.vlsi-expert.com
match as compare to
conjugate match. The gain at lower signal
levels are 1 dB or so less in power match
as compare to conjugate match.
www.vlsi-expert.com
 Maximum linear output power (point A and
A’) and power at 1 dB compression point
(point www.vlsi-expert.com
B and B’) is 2 dB higher in case of www.vlsi-expert.com
power match as compare to conjugate
match.
www.vlsi-expert.com

The improvement may vary between 0.5 dB – 4 dB. Fig. PA characteristics for conjugate match and
power match [1]
www.vlsiexpert.com | www.technoready.com
OBTAINING POWER MATCH
A load pull measurement setup consists of Device Under Test
(DUT) with some form of calibrated impedance tuning at its
www.vlsi-expert.com www.vlsi-expert.com
www.vlsi-expert.com
output. The input may or may not be tunable. Input is mainly
matched to boost the power gain i.e. conjugately matched at
the desired frequency.www.vlsi-expert.com

Using this output tuner, different impedance are shown to the


www.vlsi-expert.com
DUT over the smith chart region and its output parameters Fig. Typical load pull data [1]

such as output power, efficiency etc. are measured.


www.vlsi-expert.com www.vlsi-expert.com
This process of measuring the output parameters by changing
the impedances shown to DUT using tuner is known as load
www.vlsi-expert.com
pull measurement.

www.vlsiexpert.com | www.technoready.com
PA Classes : Class A

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
PA Classes : Class B

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
PA Classes: Class C

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
SUMMARY
Class of operation
www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com
depends on the conduction
angle (time for which the
www.vlsi-expert.com

device conducts current)


www.vlsi-expert.com
Class-A and push-pull
www.vlsi-expert.com
Class-AB are inherently www.vlsi-expert.com

more linear than class


www.vlsi-expert.com

C,D,E,F,J
www.vlsiexpert.com | www.technoready.com
SUMMARY

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Design Steps for Power Amplifier
Plot contours of constant power output on a smith
chart as a function of ΓLP with transistor conjugately
matched at www.vlsi-expert.com
its input- These are called Load-pull
www.vlsi-expert.com www.vlsi-expert.com
contours.

Select suitable ΓLP depending


www.vlsi-expert.com
on Desired P1-dB

Calculate ΓIN Using bilateral design formula based on


ΓLP instead of ΓL
www.vlsi-expert.com
Use conjugate matching at the input i.e
Γ𝑠𝑠𝑠𝑠 = Γ𝑖𝑖𝑖𝑖 ∗
www.vlsi-expert.com www.vlsi-expert.com
Use small signal S-parameters to check
stability www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Doherty Power Amplifier

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

 If an auxiliary transistor is introduced that provides gain only when the main transistor begins to compress,
www.vlsi-expert.com
then the overall gain can remain relatively constant for higher input and output levels.
www.vlsi-expert.com

www.vlsi-expert.com
 If the voltage swing at X is large enough to drive M1 into the triode region, then it is likely to drive M2 into
the triode region, too.

www.vlsiexpert.com | www.technoready.com
Doherty Power Amplifier

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
Class-F Waveform

www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com
www.vlsi-expert.com www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

Thank You www.vlsi-expert.com

www.vlsi-expert.com www.vlsi-expert.com

www.vlsi-expert.com

www.vlsiexpert.com | www.technoready.com

You might also like