Project 13 - Impedance Matching Design Using Lumped and Distributed Implementations in ADS (October 2013)
Project 13 - Impedance Matching Design Using Lumped and Distributed Implementations in ADS (October 2013)
Stephen Watt
Due: 10/25/13
Partners: Charles Green, Brock Baily
OBJECTIVE
Two relatively simples techniques for narrow band impedance matching are L-section matching networks
and single stub matching. The student explored the design, simulation, and layout of these matching
circuits using the Smith chart and Agilent ADS tools.
= ( ) + (
) = . + . []
(. )
where is equal to 2*(1.1GHz). This load is normalized in a 50 system and plotted on the Smith chart.
= . + .
Part B
See the attached hand calculations for all eight possible methods of lumped matching. For the lumped
elements matching the first reactive component is used to bring the normalized impedance to the r=1 or
g=1 circle (depending if using a shunt or series element). Then another reactive component is used to
bring the normalized impedance to the matched condition (1+j0).
To accomplish the lumped element L-C matching of the antenna load, the Smith Chart and component
tuning tools in ADS were used. Using the Smith Chart tool the four L-C configurations and the
approximate component values were determined. After incorporating the L-C matching, the tuning tool in
ADS provided much more precise component values needed to achieve a matched condition.
Part C
Four tuning configurations exist for completing single stub matching. An open circuit or short circuit stub
can be used, and each of these can be used in series or shunt. The realizability of the series and shunt
elements will be discussed in Section C. For the stub tuning of the antenna load the Smith chart hand
calculations were completed to determine the stub lengths and distance from the load. The stub distances
and lengths were gathered in a table and ADS simulations were completed to measure the matching
circuit performance.
SIMULATION SET-UP A
SIMULATION SET-UP B
Methodology
Results
|0|
ZIN/Z0
Series C (pF)
Shunt C (pF)
0.003
1.006 + j0.000
3.74
6.08
0.001
1.000 j0.001
3.73
6.075
|0|
ZIN/Z0
Series L (nH)
Shunt C (pF)
0.001
1.001 + j0.000
5.61
8.88
0.001
0.999 + j0.001
5.615
8.88
|0|
ZIN/Z0
Shunt C (pF)
Series C (pF)
0.001
0.998 + j0.000
9.29
35.4
0.001
1.000 j0.001
9.29
35.39
|0|
ZIN/Z0
Shunt L (nH)
Series C (pF)
0.001
1.002 + j0.000
2.26
4.45
0.0002
1.000 + j0.000
2.253
4.455
Analysis
While simulating using the COTS components from the Murata and Coilcraft libraries, the precise
component values required for the matching circuits were not always available. For future circuit design
care would be taken to ensure all of the matching components used are commercially available and in the
appropriate ADS libraries. The board layouts will need to incorporate enough room on the input trace for
a connector (such as an edge-mount SMA) to be soldered in place. The ground plane should be as large as
possible with vias that maintain a spacing of 0.1g between them.
The L-C matching is a narrow band matching technique and is used for relatively low frequencies. This
is due to the self resonant frequencies of the reactive components, which make the matching circuit less
than ideal for higher frequencies.
SIMULATION SET-UP C
The four possible stub tuning configurations are identified in Figure 21. The distributed series matching
circuits were calculated using series stub elements. A microstrip series stub is not physically realizable so
future calculations will be completed for the shunt stub solutions. As indicated on the attached Smith
chart, there are two possible solutions for the series stub configuration.
Distance from ZL
OC Stub Length
Electrical Length (, )
0.142, 51.12
0.056, 20.13
Physical Length (cm)
3.870
1.530
Physical Length on FR4 (cm)
1.870
0.738
Table 5. Series stub location A from Smith chart
SC Stub Length
0.363, 129.6
9.890
4.769
Distance from ZL
OC Stub Length
Electrical Length (, )
0.228, 82.08
0.453, 163.08
Physical Length (cm)
6.21
12.35
Physical Length on FR4 (cm)
3.0
5.96
Table 6. Series stub location B from Smith chart
SC Stub Length
0.203, 73.08
5.53
2.67
Analysis
Single stub matching is a narrow band matching technique. The stub matching is considered as distributed
matching and is used for relatively higher frequencies. For future microstrip stub matching design only
shunt stub elements will be considered since this configuration is physically realizable. A series stub
element is physically realizable for a coaxial circuit but not for a microstrip circuit.
Matching Circuit
Series C, Shunt C
Soln A, OC Stub
|0|
0.07
0.05
RL(dB)
-53.19
-59.91
Pdel(dBm)
37.2
38
Pdel(W)
5.25
6.31
Question 2
Using a shunt capacitor and a quarter wave transformer the matching of the antenna load at 1.1 GHz was
achieved. First the shunt capacitor impedance value was chosen to be equal and opposite magnitude to the
reactance of the antenna load. Next the impedance of the quarter wave transformer was calculated using
= 0 = 63.25
with Z0 as 50 and ZL as our antenna impedance. The plot of reflection coefficient versus frequency
displayed the achievable bandwidth as 24 MHz or 2.18% when max=0.05.
Question 3
The most suitable matching circuit will depend on the frequency of operation, bandwidth, physical
realizability, and manufacturing costs. The quarter wave transformer, L-section L-C matching, and single
stub matching circuits were considered during this analysis. For higher frequency circuits, stub matching
will be more suitable than L-C matching due to the parasitics and self-resonant frequency of the lumped
components. The stub tuning technique may prove to encompass larger physical dimensions than a
lumped element solution, which could translate into higher manufacturing costs. The quarter wave
transformer has a higher bandwidth than the other matching techniques analyzed here but the circuit may
cover larger physical dimensions than a lumped or stub tuning circuit.