RF and Microwave Component Design A PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 11

Summary of RF Measurement Labs EERF 6396

Sujeet Patole

Spring 2013

1 Introduction

This report summarizes design of several Microwave components which were designed as the part of the
course. The designs were carried out using AWR Microwave Office and AXIEM EM simulator. The milled
designs were verified in LAB using VNA, spectrum analyzers and other commonly used tools.

Microstrip lines on either FR-4 or Duroid substrate were used as basic building blocks of these designs
which operate in GHz frequency range.

The report is organized as follows: Each section is dedicated to one component. Subsections discuss theory
of design and lab results along with common applications.

2 Resonators

2.1 Design of Resonators


• Design of Quarter wave resonator: Consider input impedance of terminated transmission line,
Zl + jZo tan βl
Zin = Zo (1)
Zo + jZl tan βl

For a quarter wavelength line which is open at one end (Zl = ∞, βl = 900 ) thus, Zin = 0 i.e. QW
transformer inverts impedance. Hence, 50Ω through transmission line with quarter wavelength open
stub in shunt will be shorted to ground at resonance frequency giving return loss of 0 dB and very high
insertion loss. Using TX line tool in AWR MWO with electrical length of 900 , an open circuit stub is
constructed.

1
Figure 1: Schematic of quarter wave and parallel resonator

• Design of two stubs Parallel LC resonator: Again consider input impedance of terminated transmission
line (1), For a open circuit stub,
Zin = −jZo cot βl (2)
Thus when electrical length is less than 900 stub acts as a capacitive and when it is greater than 900
it is inductive. We can match two stub lengths so that resonance is obtained at particular frequency.

2.2 Implementation of resonators

Figure 2: Circuit Photographs of quarter wave and parallel resonator

2
Figure 3: Lab Measurements of quarter wave and parallel resonator

2.3 Applications of resonators

Resonators can be used as a filter at microwave frequencies. Depending on the narrowband or passband
applications the value of Q factor may be decided. Narrowband system has high Q factor, while broadband
systems low.

3 Microwave Filters

3.1 Design of the filters

2 2
Filters are two port passive reciprocal devices. Ideal filters are lossless i.e. S11 + S21 = 1. Thus we can
2
write Pout = (1 − |Γ| )Pin . Aim of the filter design is to control reflection coefficient Γ as a function of the
frequency. Thus at the desired passband a match is created between input and output, however at the other
frequencies a filter creates an impedance mismatch between input and output so as to have low Γ and less
Pout . Insertion loss of filter is given by,
Pin
IL = 10 log10 = −10 log10 (1 − |Γ|2 ) (3)
Pload
Two techniques that are followed for designing microwave filters is series-shunt transmission lines and
stepped impedance transmission lines.

• Series Shunt Impedance Filter Series lines act as a short circuit stubs. Thus, using microstrip lines
with the characteristics impedance Z0 we can find the length of the required stub for given Zin = XL .
Zin = −jZ0 cot βl (4)
Parallel lines act as a open circuit stubs. Thus, using microstrip lines with the characteristics impedance
Z0 we can find the length of the required stub for given Zin = XC .
Zin = jZ0 tan βl (5)

3
• Stepped Impedace Filter For the microstrip lines when βl <= 450 . Following approximations hold
good. Here, Zh and Zl are highest and lowest impedances possible.
LR0
βl = (6)
Zh
CZl
βl = (7)
R0

Figure 4: Stepped impedance filter

3.2 Implementation of filters

Filter coefficients were obtained from design charts of the Chebyshev (Series shunt design) filter and But-
terworth (stepped impedance design) filter. 4 GHz low pass filters were constructed using the equations
discussed in previous sections.

Figure 5: Photos of stepped impedance and series-shunt filters

4
Figure 6: Lab measurements of stepped impedance and series-shunt filters

3.3 Applications of filters

Microwave filters control the frequency response of the system. They are used in communication system at
RF transmitter to bandlimit the transmitted signal and in RF receiver front end to limit down converted
bandpass signal.

4 Microwave Power Dividers

4.1 Design of Wilkinson power divider

Wilkinson power divider is a three port network. From the scattering matrix conditions, a three port network
can not be lossless, reciprocal and matched at all ports simultaneously. For the physical realization of
the circuit one of these condition should be relaxed.

For example T-junctions are lossless, however the junction is not matched at all ports. There exists a
mismatch looking into output ports for the matched input ports. It is reciprocal and its scattering matrix
shows sum of columns=1.  √ √ 
0√ −j/ 2 −j/ 2
Slossless = −j/√2 S22 S23  .
−j/ 2 S32 S33
Output lines can be matched using quarter wave transformer. Resistive divider is another type which is
matched at all ports. It is reciprocal device. However, it is a lossy as V2 = 1/2 × V1 . Hence S12 = S31 = 1/2.
 
0 1/2 1/2
ST = 1/2 0 1/2 .
1/2 1/2 0

Wilkinson power divider is same as lossless power divider except a resistor between port 2 and 3.

5
Figure 7: Wilkinson power divider

Wilkinson power divider is lossless when output ports are matched. Thus, while dividing input power
wilkinson power divider is lossless. When output ports are not matched power is dissipated in resistor and
it is lossy. Port 2 and 3 are isolated i.e. S23 = S32 = 0. Its S matrix is given by
 √ √ 
0√ −j/ 2 −j/ 2
Swilkinson = −j/√2 0 0 .
−j/ 2 0 0

Wilkinson power divider is a reciprocal device. It equally splits input signal into two equal power and equi-
phase signals and can combine them in opposite direction.

4.2 Implementation of Wilkinson power divider

Wilkinson power divider is designed for the 3dB power split. Design is carried out using MWO and AXIEM.
The design is fabricated on Rogers 5880 Duroid board and is tested in lab.

Figure 8: Circuit photograph and lab measurements of 3dB power divider

6
4.3 Application of Wilkinson power divider

Wilkinson power divider can be used for splitting as well as combining power networks.

5 Microwave Couplers

5.1 Design of Couplers

Couplers are four port device as shown in figure below,

Figure 9: Block diagram of coupler

Power from port 1 is transmitted to port 2 (direct or through port) and coupled to port 3. Port 1 and 4 are
isolated and same is true for port 2 and 3. Couplers are reciprocal deivices. 3 important parameters are
used to characterize coupler.
P1
Coupling = C = 10 log10 (8)
P3
Ability to coupler separate forward and backward wave is given by,
P3
Directivity = D = 10 log10 (9)
P4
P1
Isolation = I = 10 log10 (10)
P4
I = D + C (dB) (11)
Unlike a 3-port device, a matched, lossless, reciprocal 4-port device is physically possible. Two general
solutions exist for this problem. The first solution is symmteric gives S matrix
 
0 α jβ 0
α 0 0 jβ 
Ssymmetric =  jβ 0
.
0 α
0 jβ α 0

This is know as quadrature hybrid or 900 coupler (900 phase shift between input and couple ports or two
output ports or between input and direct port due to presence of λ/4 lines). Another solution gives anti
symmetric matrix with 1800 hybrid coupler. One realization of quadrature hybrid coupler is branch in line
coupler which consists of consists of two parallel transmission lines physically coupled together.

7
Figure 10: Branch in line (quadrature hybrid) coupler

Qudrature hybrid is 3 dB coupler. Other coupling ratios are possibel using coupled line design. Coupled
lines have two plane reflection symmetry. Like other quarter wavelength designs it is a narrowband device.
Coupled lines are physically separated. Spacing between quarter wavelength lines decide coupling ratio.

Figure 11: Coupled line coupler

5.2 Implementation of Couplers

Figure 12: Circuit photograph branch line and coupling line coupler

5.3 Application of Couplers

Couplers are used to get better return loss at input and output ports. This technique is used in balanced
power amplifier, PIN diode based mixers, etc.

8
6 Microwave Amplifiers and Noise Figure Measurement

6.1 Theory of of amplifier design

Figure 13: Amplifier with IO Matching networks

Aim of the matching network and amplfier design is to couple maximum power from input to gain element
(i.e. transistor in this case) and to deliver maximum power from the load to gain element. One way to
achieve maximum coupling at the input would be to modify source impedance ZS to match the transistor
characteristics. However, this results in large loss across this impedance, Hence only fraction of input power
is only coupled. To solve this problem, input matching network with ΓS = Γ∗in and output matching network
with Γout = Γ∗load is constructed.Whenever, source and load impedances are conjugately matched, maximum
power transfer is obtained between the stages.

6.2 Theory of noise figure measurement

For the Noise figure measurement we note that output noise of the amplifier is given by,

No = G(Ni + Nadded ) = F KT B (12)

Where Nadded = (F − 1)KT B is amplifiers noise contribution. Substituting this expression in above formula
we can find the Noise figure of the amplifier. Following is the assembly for the noise figure measurement.
Spectrum analyzer is used to measure noise floor. Note that B in (12) is the Resolution Bandwidth of
Spectrum Analyzer.

Figure 14: Noise figure measurement

9
6.3 Impementation of Amplifier

Microwave amplifier operating in the frequency range of 1-2 GHz with its input and output matching network
is designed.

Figure 15: Circuit photograph amplifier and lab readings

6.4 Applications of amplifier

Microwave amplifiers have wide applications in the communication industry. They are used to amplify weak
received signal as well as to boost the signal before transmission. Applications and use of amplifier vary
according to the gain and its frequency range.

7 MMIC Amplifier Design

MMIC amplifier with feature size of 12µm was built. Capacitors and Inductors are realized using transmission
lines. Resistors are built using square blocks (50µm × 50µm) of T aN (feedback loop) and square blocks
(100µm × 100µm) of AuGeN i (input line). DC Blocking capacitors at the RF ports and Bypass capacitors
for the DC power supply were used.

10
Figure 16: Layout of MMIC Amplifier

8 Patch Antenna Design

Refer the following link,

http://www.utdallas.edu/~sujeet.patole/patch.htm

11

You might also like