ELECTENG 721 Radio Engineering

ELECTENG 737 Advanced Radio Engineering

Laboratory 1 — Microstrip Filter Design Using

CST Microwave Studio®

This laboratory involves the design, implementation and characterisation of a microstrip

filter using the numerical electromagnetics simulation software package CST Microwave
Studio®1 [1]. The specification of the microwave filter is identical to that covered in
the course notes [2], but whereas a coaxial realisation was considered therein, in this
laboratory we will consider a microstrip realisation.

Please read and follow the instructions set out in this document carefully, as many mean-
ingful results can be obtained from CST Microwave Studio® but care must be taken to
ensure that this software is driven correctly. Marks will be awarded based on the
answers you provide in the Canvas quiz ‘Laboratory 1’. The questions in this
Canvas quiz are repeated in this document and are shown in a red box .

It is also strongly recommended that you review the coaxial filter case study
in the lecture notes [2] You should be able to answer Questions 1 to 7 in this
document in advance of your scheduled laboratory session.

1 Introduction

CST Microwave Studio® is part of CST Studio Suite® , which is a collection of three-
dimensional modelling tools used in the simulation of electromagnetic problems2 . A major
capability of CST Microwave Studio® is the simulation and optimisation of high-frequency
electromagnetic systems. The flagship numerical solver used by CST Microwave Studio®
is based on a time domain integral formulation of Maxwell’s equations, known at the
Finite Integration Technique (FIT) which is similar to the well-known Finite-Difference
Time-Domain (FDTD) technique (e.g. [4, p74ff]). This technique involves dividing the
structure to be analysed into a number of hexahedral cells (which are often referred to as
mesh cells). Upon specifying a suitable time domain excitation signal, numerical results
for electromagnetic field quantities such as E, H, power flow, input impedance, etc can be
generated. In general, increasing the number of mesh cells for a given geometry increases
1
This version of the laboratory manual is intended to be used with CST Studio Suite® 2024.
2
CST Studio Suite® is a component of Simulia, which includes software for range of additional domains
(including mechanical/structural simulation) and which in turn is marketed by Dassault Systems [3].

1
the accuracy of the numerical solution. However, this comes at the expense of increased
memory requirements and computation time. In CST Microwave Studio® the mesh is
generated automatically upon specifying the frequency range of interest. This default
mesh is usually sufficient for most problems.

Once the simulation has completed, field quantities of interest can be represented graph-
ically and even animated. This makes CST Microwave Studio® a very powerful electro-
magnetic analysis tool for engineers, as well as a useful aid in the teaching of electromag-
netics — a subject that, due to a lot of vector mathematics, produces results that can
often be challenging to visualise.

Apart from the time domain solver, CST Microwave Studio® includes a number of other
solvers including frequency-domain and eigenmode solvers. In this laboratory only the
time-domain solver will be used. Further information on computational electromagnetic
techniques in general can be found in a number of texts, e.g. [4, 5].

2 Microstrip Fundamentals

A perspective representation of a microstrip transmission line is shown in Fig. 1 and

a cross-sectional view in Fig. 2. In order to design a microstrip transmission line, it
is necessary to obtain expressions for the characteristic impedance Z0 (Ω) and guide
wavelength3 λg (m).

2.1 Characteristic Impedance Z0

The characteristic impedance Z0 of a microstrip transmission line is a function of its width

w, the dielectric substrate height h, the thickness of the metal (cladding) t, the relative
permittivity of the substrate ϵr and is given by [6]
  s 
2  2
42.4 4h  14 + 8/ϵr 4h 14 + 8/ϵr 4h 1 + 1/ϵr 2 
      
Z0 = √ ln 1 + + + π
ϵr + 1  w′ 11 w′ 11 w′ 2 

(1)

where w′ is the equivalent width of a thin strip given by

 
1 + 1/ϵr  4e
w′ = w + t

ln  r  . (2)
2π 
t
2 
1/π
 2
h
+ w/t+1.1

It is not possible to solve (1) for w′ (and then (2) for w) in closed form, although it is
straightforward to solve for these parameters numerically. Alternatively, the graphs in
Figs. 11 and 12 can be used to determine the microstrip line width w for a required Z0
(results for Z0 are provided for three different substrate thicknesses h and assume ϵr = 2.2
and 35 µm cladding).
3
On microstrip transmission lines, the guide wavelength is different to the free-space wavelength as
the fields propagate in both the dielectric substrate and air.

2
Figure 1: Microstrip transmission line — perspective view. The conductors (microstrip
track and ground plane) are shown in brown, and the dielectric substrate in blue.

Microstrip track
Substrate
w

t
Ground plane
Relative permittivity ϵr h

Figure 2: Microstrip transmission line — cross-sectional view.

3
 Parameter Description Value
ϵr Substrate relative permittivity 2.2
h Substrate thickness 3.175 mm
— Cladding material Copper
t Cladding thickness 0.035 mm

Table 1: Key specifications of Rogers RT/duroid 5880 high frequency laminate.

2.2 Guide wavelength λg

Energy propagates on a microstrip transmission line in both the air region above the
dielectric substrate, and in the substrate. This inhomogeneity of regions means that the
guide wavelength is also a function of the dielectric properties of these two layers. It can
be shown that the guide wavelength λg for a microstrip transmission line is
c
λg = √ (3)
f ϵeff
where the ‘effective’ relative permittivity of the air/substrate ϵeff is given by [7]
 
ϵr + 1 ϵr − 1  1 
ϵeff = + r
  .
 (4)
2 2  h
1 + 12 w

Given the desired characteristic impedance Z0 , equations (1) and (2) (or alter-
natively the graphs in Figs. 11 and 12) can be used to determine the microstrip
line width w. Once w is known, equations (3) and (4) can be used to determine
the guide wavelength λg .

3 Microstrip Low-pass Filter

The low-pass filter equivalent circuit and its transmission line representation are shown in
Figs. 3 and 4 respectively. Similar to the coaxial realisation described in [2], the microstrip
realisation will consist of a number of series connected transmission line elements, as
shown in Fig. 5. The filter will be implemented on Rogers RT/duroid 5880 high frequency
laminate [8], as specified in Table 1.

1. The first step is to determine the widths w0, w1 and w2 corresponding to the charac-
teristic impedances Z0 , Z01 and Z02 respectively. Equations (1) and (2) can be used
for this purpose (although this will require a numerical solution), or alternatively
using the graphs in Figs. 11 and 12 (recommended). Please do not write on
the graphs on pages 15/16 — separate copies of these graphs have been
provided for you to write on if required.

Question 1: For h = 3.175 mm, t = 0.035 mm and Z0 = 50 Ω, calculate w0.

Question 2: For h = 3.175 mm, t = 0.035 mm and Z01 = 176 Ω, calculate w1.

4
 L1 /2 = 17.6 nH L1 /2 = 17.6 nH

C2 = 2.55 pF

Figure 3: Low-pass filter equivalent circuit.

Z0 = 50 Ω Z01 = 176 Ω Z02 = 19.7 Ω Z01 = 176 Ω Z0 = 50 Ω

l1 l2 l1

Figure 4: Low-pass filter transmission line representation.

y l0 l1 l2
z

x
l1 l0

w0 w1 w1 w0

w2

Coaxial feed locations (through ground plane)

Figure 5: Low-pass microstrip filter — top view.

5
 Question 3: For h = 3.175 mm, t = 0.035 mm and Z02 = 19.7 Ω, calculate w2.

2. Now that the widths w0, w1 and w2 of the different transmission line sections have
been calculated, the lengths l0, l1 and l2 must be determined. The length of the
input and output tracks (l0) is not critical and is set to 20 mm. The lengths l1 and l2
must be l ess than λg /4 at the maximum frequency of interest, 5 GHz. These lengths
can be calculated by determining λg from (3) and (4), using the values calculated
for the widths w1 and w2 in Questions 2 and 3 (and using the microstrip parameters
in Table 1). (You can assume f in (3) is 5 GHz.)

Question 4: For Z01 = 176 Ω and ϵr = 2.2, calculate ϵeff .

Question 5: For Z01 = 176 Ω and ϵr = 2.2, calculate l1 = λg /4.

Question 6: For Z02 = 19.7 Ω and ϵr = 2.2, calculate ϵeff .

Question 7: For Z02 = 19.7 Ω and ϵr = 2.2, calculate l2 = λg /4.

Now that all of the physical dimensions of the filter have been determined, modelling the
design in CST Microwave Studio® can commence.

3.1 Implementing the Filter in CST Microwave Studio®

A design template (in which the filter has been mostly implemented) is provided on Canvas
(filename ustrip-1.cst). This file needs to be downloaded and copied to a convenient
location on your computer. (It is recommended that you create a folder CST in your
OneDrive Documents folder, and then copy the template file to this location.)

The next step is to start CST Microwave Studio® from the Windows desktop:

1. Click , and then search for CST. Click on CST Studio Suite 2024.

2. A dialogue window will appear acknowledging the university license agreement. Click Ok .

3. A dialogue window may appear offering to show a quick introduction video. If this appears,
click No .

4. On the screen that appears, click Open from the list on the left hand side, navigate to
the template file that you have downloaded from Canvas, then click Open . The filter
template will open and you should see the screen as shown in Fig. 6.

6
 Figure 6: CST Microwave Studio® microstrip filter implementation — initial screen.

The menu options on the top of the window and their associated icons are used to control
the software. The Navigation Tree on the left-hand side of the window is used to access
the various components in the design, materials used, details of excitation(s) and also the
results (obtained once the simulation runs have been completed). The Parameter List
at the bottom of the window is used to define parameters (typically dimensions) which
are used to create the design (note that all dimensions listed are in mm). It is important
to note that CST Microwave Studio® is a parametric modeller, and that the parameters
names as listed can be used to directly specify the dimensions, etc of the various geometric
entities in the design. This means that any changes to the parameter specifications can
be used to automatically update the design! The Messages/Progress pane on the lower
right-hand side of the window provides information on the simulation process once it is
running.

The filter design itself is displayed in the central pane. The large grey box is the ground-
plane (modelled as PEC, as are all metal objects in this design) and the blue material is
the dielectric substrate. The various microstrip sections are shown in light grey on top of
the substrate. The input and output (Z0 = 50 Ω) microstrip traces are fed from coaxial
transmission lines from the underside of the groundplane.

Now is a good time to view the key features of the filter implementation.

1. To observe the coaxial feed locations, select the menu item View → Mouse Control →
Rotate. Click in the main part of the window and rotate the view until the underside
of the groundplane can be seen. (Note: The view can be re-centralised and scaled to
the central pane size at any time by pushing the Spacebar .) The input and output feed
locations (termed Ports in CST Microwave Studio® ) are shown in red.
2. To better visualise the cross-section of the design, select the menu item View → Sectional
View → Cutting Plane. Make sure that the cutting plane Normal → X and Position →
0. You should see the cross-section through the design in the x = 0 plane. Use the mouse
wheel to zoom in to the coaxial port for z > 0, and you should see that there is no centre

7
 conductor in the coaxial feed line. This centre conductor must be created prior to
running the simulation.

3. To create the missing centre conductor, select Modeling → Shapes → Cylinder, and press
Esc to bring up a dialog box.

4. A dialog box will appear on which the parameters defining a cylinder (to represent the
missing centre conductor) will be entered. We will make use of the parameters in the
Parameter List in this step — both to position the center conductor in the xz-plane, to
define its vertical position and length, and its radius (defined by parameter a coax). Enter
the following:
Name → Feed 2 centre
Orientation → y
Outer radius → a coax/2
Inner radius → 0.0
Xcenter → 0
Zcenter → l2/2 + l1 + l0 - b coax/2
Ymin → -d gp
Ymax → d s + d c
Segments → 0
Leave Component set to component1 and as this object is assumed to be a perfect con-
ductor select Material → PEC.
Click OK . You should see the new centre conductor appear within the coaxial feed.

At this point, the implementation of the basic filter structure is complete. At this time it is
probably a good idea to save your project, so from the File tab select Save.

3.2 Port Specification

Ports are used to define the input and output of the filter structure. These have been pre-
defined in the filter template, and can be examined by selecting Ports → Port 1/Port 2 in the
Navigation Tree (rotate the structure to observe the underside of the ground-plane if required).

It is now necessary to specify the frequency of excitation. As we want to observe the behaviour
of the filter in both the pass- and stop-bands, it is important that the frequency range selected
is sufficiently wide. From Simulation → Settings select Frequency, enter the following (units
of GHz assumed):

Fmin → 0.0
Fmax → 4.0

Click OK .

3.3 Investigation A — Uniform Transmission Line

While it is structurally complete, the lengths and widths of the various line elements still need
to be defined by modifying the parameter list. As an initial investigation, we will investigate
the performance of a uniform transmission line with characteristic impedance Z0 (i.e. when w1
= w2 = w0).

8
In the Parameter List, double click on the corresponding Expression cell and make the following
changes:

l0 → 20.0
l1 → 20.0
l2 → 20.0
w0 → (Enter your value of w0 (corresponding to Z0 ) from Question 1)
w1 → (Enter your value of w0 (corresponding to Z0 ) from Question 1)
w2 → (Enter your value of w0 (corresponding to Z0 ) from Question 1)

After you make changes to values in the parameter list, you will see a message “Some variables
have been modified. Press ’Home: Edit → Parametric Update (F7)”. Press F7 to update the
main window. You should now see that all microstrip segments have the same width (w0).

By default, as part of a normal simulation run CST Microwave Studio® will produce a visual-
isation of the port modes (i.e. the fields in the apertures of the ports) and 1D results relating
to the ports (i.e. input and output reflection coefficients S11 and S22, and forward and reverse
transmission coefficients S21 and S12 respectively). It will not (by default) provide any other
field visualisations. However, in this laboratory we are interested in visualising the behaviour
of the fields at frequencies in the pass-band (at a frequency of 1 GHz) and stop-band (at a
frequency of 3 GHz). We achieve this by adding Field Monitors using the following steps:

1. From the Simulation tab select Monitors → Field Monitor which will bring up the monitor
dialogue window.

2. Enter the following (for the pass-band monitor):

Type → E-Field
Label → (leave as default)
Specification → Frequency list (f0; f1; ...) with a value of 1.0

3. Click OK .

4. Repeat this process to add a second field monitor at the stop-band monitor frequency
(3 GHz)

We are now ready to perform the simulation.

3.3.1 Running the simulation

1. From the Home tab select Simulation → Setup Solver (click the small arrow at the
bottom of the icon) → Time Domain Solver

2. Now go to the Simulation tab and select Solver → Setup Solver

3. Enter the following:

Solver Settings → Accuracy → -40
Adaptive mesh refinement → Adaptive mesh refinement → Deselected

4. Click Start to commence the simulation.

9
   ǸȀƂȍ ȁȍ Ȃȍ  šȍ Šȍťǥ
ȍ ƗǚȍƘȍıœȍȃȍ 

 ĐȍȄŘȍ ¿ÒǒèȍĂčƕċÝȍ5%ÀăȍÆĄÓȍ ŕĜŝÁȍ KĎźÂř
Ǜȍ ȅŖȍ È
 
5
ǃ³ȍ ÜČÞIJÉłƃȍ ďȍ
·ȍ ´ȍ 

1 — E
field.
ÑđĒȍ
glI $tI (ŠIpgzX 2s°o=¤{ ,š¥½,?IššrX 7^IºÂ

Ż
ǓŦxȍ ēż
ǘŭȍ  
 ƒœDŽ
ƙx Dzȍ )’Wąȍ ȍ qz ȍ
ȍJȍ (;ŧ< ȍœȍ

Figure





7: Port
“”ƒ·œÂ  !dKmD =¦Â²”›ƒ‘Â
Ȇȍ ȇȍ Ȉȍ åȍ  kȍ
Ŀ ž_;?ȍ ˆŤ
ǜȍ 5%ǦNjȍ

 5isIÂ ȉȍ+
<ȍ
ĔǭNJ7 Ũȍ
Ęȍ #L§EœÂ ˜ 3ƍ ȍ
,ša¤a„}Â

N™
b +?ȍ NT
ȍ
)
ǰĆȍ 0ÔðFȍ
+ ǧ–ȍ
 (;=8ȍ3
%
^ ž  īi•8ȍƼƵȍĬ
.
3 ?ȍń %ȍ
ÊȊ½

qz ȍ
ȋƚ
lȍ  +)Ŋ Ǿ ,q=MÂÀ „
!ǹ 5
•’Yŷȍ NŞ
Ɯ“ ȍ 3Ɲ
;Y ȍǿȍˆƅćȍ Ŏ
 ȍ
 LƄƛ Xȍ
ȍ ĝ£¨ȍ „< ȍ
{bt9¦IÂ

 
 5=š‹=–I{¦Â "cInFžÀ Á %~š¦={¦=zI±šÂ

< ȍ [@”¤ȍ  ȍŏfǁȍ 4KC«dƒ€:o 7dK¸Â Ğ ƞ [ȍ+
ȍ Œƿ ǔȍ ğ ũȍ
*=µ_Y=¦`‚|Â6‘II 
c ǡǐ >Ǻ—ŗ^ôȍ


Â3J=’C]Â
†yŒ†|N{­Â ĕâƱȍ
k WÂ Â
 >‚uŒ‚{Iz®Â wïoÐ
 ĭ$">
 Â 4
 ƒ7
ijȍȍ Â (
- vEȍ
˜  hȍ Ġ$§
 Â 
 vȍ
 hȍ ėƹR‘"Džȍ"
!ȍ
   P nj  tEȍ
O ªƟȍ|
`

tȍ
¾¿¾Â Â


ƭǬ/> ȍę/Ya
  `uO{š`‡zš


 Ơ
!ȍP
§ Eȍ

 (
$Ǟ  ȍŋǨ-


 L Sȍ6%Ǒ ȍ
µ2ȍ "Ÿǟȍ
ëwȍ


   }ª- a7 ȍ6 9
ȍ ìȍ

 L Sȍĸ ¡Ǚȍ
u  ȍ¯®ȍé¸ Dȍ
  ui ȍ¯®õ¹
ȍ
O "
9ȍ
! ȍġǎǏ!ȍ4#! ǝ#$
gȍg2ȍ
à  2ȍ ǍǀŚŪǕ
¶2ȍ 4Ɓ
ĪǻË°ȍ ȍJȍŅǩư
Uȍ *#\bȍ 9
ȍ
Uȍ 6j*
7/¢ȍ
ŵȍȍ 6j

/¢ȍB{Mºȍ
Vȍ Mů  !RȍĴ/>
Rȍ
V¬­ȍ I

-ȍ
Vȍ ȍ I

-ȍB{M»ȍ
UĨȍ *#ǯ Ÿȍ
Ʈħȍ Kƺ9dzȍ
Ưȍ *#$ȍ ű#›ƻ‹`Ɔ"ȍ
ĚǼ̱ȍ Jàö|ȍņ %=`ȍ
Ȍȍ¬­ȍ *\bȍ4Sȍ
ěĩȍ "$aơȍ



Ƈȍ ˜²Ƣȍ ~džǂǪfȍ įǵȍ




Ĉȍ Ķ² ųƣȍ
Ŷȍȍ ƽ ȍ
ŀŽ
ȍ
) ȍfȍ

őKĹȍ
¾ ȍ }Í~ Ƥ ķƐWūȍ ĵmȍ ûÕñ1ürȍ
gȍ±ȍ (
\Ǡ  ǖ P
¨ȍ€Öȍ 0÷m''rȍ
ΰȍ Ő
ś I
ȍ nÿ1ȍsáȍ
s×F1Dþ1Ïúȍ
)
e“ȍBƥ ¼ȍ DênFȍĖäȍ
?@²‘9?¼Â

øĢȍ 3B\Qv;¬gA €ȍ îĥȍ† Ǘǫ–Ǥ‰ŁZ¡ȍ¥ƈŸƶ
]ȍ óæãùģȍҎ AƦ@]Š £^ȍĺZŒŮ¥Š Z$@Ƨ‰ž—ȍ


Ǭ  DZȍ )QŸ <YRÂ
, , ȍ, C , ,ǽ C &&&&&&
C ßÚȍ
  
-;•<yP¨Q•Â'` 8Â

 ƒ
™ ȍ O
% ȍ Ĥ Ʒ  TX



Ž•Ižše{Â
HȍGØȍ Gkȍ T ȍĽȍdƉ :ȍ 3¦Q=G»Â ¦=¦QÂP|Q”[ÂB–`ªQ”`‚|ÂxP¤¡‚nµQ”ÂV|` ]QG¢³BAQ£¢U´oq»Â
 © Hȍ'oȍ 'Ûȍ ‚ƎŬȍȍdƊ8:ȍ
‡c..]?ŲAŔ«ȍŴƸƓſ Ţȍ‡ .ƾšŒ
[«ȍŃA
›ǣȍĦǮ
=c
 #šlȍ
 © ĉȍ'Ùȍ Gpȍ X
‚Ƒ ȍòȍdƋ8_:ȍ
 QŹ Hȍ0ȍ 0ȍ   ȍ=
ȍǢ:”.Ɩ ȍ 0I=¯|Y=–=yIª”fC ”S´n¦šÂU‚—Â1´|Â&ÂÂ
 Q Qe¦ Ċȍȍ íȍ
Figure 8: Port 1 — H field.
ŌŜ @LJ
ȍ W ¦‘ȍ e ƌ ȍ
 ¹9™`|Y ‚AA±–‘PHÂ

.;•;wQ©Q”Â'h¢¬Â †ŰAƬ_ȍ +<¶jY:¤ˆ–  )Q¢¢;ZR /•‰Y•TÂ

ň‹ţǴȍ


  Ƿȍ ƨƔȍ
  ʼn
ÅƏLjyōpľȍ ĻŽ¤ƀşƪƫ yāýÄĀçȍ ļ ljƲ
Ʃȍ Ƴƴȍ ĮİǶȍ  ȍ 

5. Once the simulation completes4 , go to the 1D Results folder in the Navigation Tree (lo-
cated to the left of the main window) and select Port Signals. The results with an i prefix
represent the band-limited impulses used to excite the system (from Ports 1 and 2), and
the results with an o prefix represent the corresponding impulse responses, respectively.
These results, though not particularly useful by themselves are used by the software to
calculate S11 and S21.

6. Now navigate to the 2D/3D Results folder in the Navigation Tree, and select Port Modes
→ Port1 → e1. Rotate the view and zoom in to Port 1 on the underside of the ground
plane. The electric field in the port is displayed, and you should obtain a result similar to
that in Fig. 7 (you may need to zoom in on the port to see the structure of the field).

7. Now select Port Modes → Port1 → h1. Rotate the view and zoom in to Port 1 on the
underside of the ground plane. This displays the magnetic field in the port, and you
should obtain a result similar to that in Fig. 8. Note that this results for the electric and
magnetic fields are consistent with that of a TEM mode (as expected). Press Spacebar to
zoom out, and click on the 3D tab on the bottom of the central pane to return to the 3D
view of the filter.

At the conclusion of the simulation, you will see a warning icon (") displayed in the Messages
4

window at the bottom of the screen. This is usually just a reminder of the limitations of the educational
license, and can be safely ignored. Other warnings however, should not be ignored as they may indicate
some problem with the simulation that may impact the accuracy of the results.

10
 Figure 9: Uniform microstrip line — S11.

Figure 10: Uniform microstrip line — S21.

8. Now navigate to the 1D Results folder in the Navigation Tree, and select S-Parameters
→ S1,1 (Port 1 reflection coefficient). Select 1D Plot → Plot Type → Smith Chart. You
should obtain a result similar to that in Fig. 9. Note that the simulated characteristic
impedance in the top right hand corner of the screen is very close to 50 Ω (as expected).

9. Navigate to the 1D Results folder in the Navigation Tree, and select S-Parameters →
S2,1 (forward transmission coefficient). You should obtain a result similar to that in
Fig. 10. Note that the insertion gain of the uniform transmission line is close to zero (as
expected).

10. We will now visualise the field in the dielectric and in the space above the microstrip
track. Navigate to the 2D/3D Results folder in the Navigation Tree, and select E-field
→ e-field (f=1) [1] (corresponding to the E-field monitor at 1 GHz, for Port 1 excitation).

11. We will first examine at the behaviour of the E-field normal to the direction of prop-
agation. In the menu bar at the top of the screen, select Plot Type → Instanta-
neous/Real/Contour/Tangential, and Sectional View → Fields on Plane (click the

11
 icon)/Cutting Plane (click the icon) and Normal → Z (this defines a cut-plane at z = 0
on which the real part of the E-field tangential (to the cut-plane) will be displayed). Click
Plot Type → Animate Fields. You should zoom in to observe the fields in the vicinity of
the microstrip line.

Question 8: For the uniform transmission line, you have plotted the E-field normal to the
direction of propagation. Describe what you observe and interpret/explain your observations.

12. We will now examine at the behaviour of the E-field tangential to the direction of
propagation. In the menu bar at the top of the screen, select Plot Type → Instanta-
neous/Real/Contour/Z, and Sectional View → Fields on Plane (click the icon)/Cutting
Plane (click the icon) and Normal → X (this defines a cut-plane at x = 0 on which the
real part of the E-field in the z-direction will be displayed). Click Plot Type → Animate
Fields. As the field amplitudes in the z-direction are quite small in amplitude you may
need to adjust the scaling of the colour ramp in order to see the field variations; Color
Ramp → Min → −200 and Color Ramp → Max → 200 is suggested.

Question 9: For the uniform transmission line, you have plotted the magnitude of the E-field
tangential to the direction of propagation (i.e. the z-direction). Describe what you observe
and interpret/explain your observations.

We are now ready to examine the behaviour of the microstrip filter.

3.4 Investigation B — Microstrip low-pass Filter

The first step is to update the various line lengths and widths.

In the Parameter List, double click on the corresponding Expression cell and make the following
changes5 :

l1 → (Enter your value of l1 (corresponding to Z01 ) from Question 5)

l2 → (Enter your value of l2 (corresponding to Z02 ) from Question 7)
w0 → (Enter your value of w0 (corresponding to Z0 ) from Question 1)
w1 → (Enter your value of w1 (corresponding to Z01 ) from Question 2)
w2 → (Enter your value of w2 (corresponding to Z02 ) from Question 3)

After you make changes to values in the parameter list, you will see a message “Some variables
have been modified. Press ’Home: Edit → Parametric Update (F7)”. Press F7 to update the
main window.

3.4.1 Running the simulation

1. Go to the Simulation tab and select Solver → Setup Solver.

2. Click Start to commence the simulation.

5
You will see a dialogue box opening entitled Results May Get Incompatible With Model which is advising
that changing any of the parameters will invalidate the current set of results. You should therefore make
sure that Delete current results is selected and click Ok .

12
3. Once the simulation completes, navigate to the 1D Results folder in the Navigation Tree,
and select S-Parameters → S1,1 (Port 1 reflection coefficient). Select 1D Plot → Plot
Type → Smith Chart.

Question 10: For the microstrip filter, you have plotted the input reflection coefficient on a
Smith chart. Describe what you observe and interpret/explain your observations.

4. Navigate to the 1D Results folder in the Navigation Tree, and select S-Parameters →
S2,1 (forward transmission coefficient).

Question 11: For the microstrip filter, you have plotted a graph of the forward transmission
coefficient. Describe what you observe and interpret/explain your observations.

5. We will now visualise the field in the dielectric and in the space above the microstrip track
in the passband. Navigate to the 2D/3D Results folder in the Navigation Tree, and
select E-field → e-field (f=1) [1] (corresponding to the E-field monitor at 1 GHz, for
Port 1 excitation).

6. We will first examine at the behaviour of the E-field normal to the direction of prop-
agation. In the menu bar at the top of the screen, select Plot Type → Instanta-
neous/Real/Contour/Tangential, and Sectional View → Fields on Plane (click the
icon)/Cutting Plane (click the icon) and Normal → Z (this defines a cut-plane at z = 0
on which the real part of the E-field tangential (to the cut-plane) will be displayed). Click
Plot Type → Animate Fields.

Question 12: For the microstrip filter, you have plotted the E-field normal to the direction
of propagation in the passband. Describe what you observe and interpret/explain your obser-
vations.

7. We will now examine at the behaviour of the E-field tangential to the direction of
propagation. In the menu bar at the top of the screen, select Plot Type → Instanta-
neous/Real/Contour/Z, and Sectional View → Fields on Plane (click the icon)/Cutting
Plane (click the icon) and Normal → X (this defines a cut-plane at x = 0 on which the
real part of the E-field in the z-direction will be displayed). Click Plot Type → Animate
Fields. (You may need to adjust the scaling of the colour ramp in order to see the field
variations; Color Ramp → Min → −200 and Color Ramp → Max → 200 is suggested.)

Question 13: For the microstrip filter, you have plotted the magnitude of the E-field in the
direction of propagation (i.e. the z-direction) in the passband. Describe what you observe and
interpret/explain your observations.

8. We will now visualise the field in the dielectric and in the space above the microstrip track
in the stopband. Navigate to the 2D/3D Results folder in the Navigation Tree, and
select E-field → e-field (f=3) [1] (corresponding to the E-field monitor at 3 GHz, for
Port 1 excitation).

9. We will first examine at the behaviour of the E-field normal to the direction of prop-
agation. In the menu bar at the top of the screen, select Plot Type → Instanta-
neous/Real/Contour/Tangential, and Sectional View → Fields on Plane (click the
icon)/Cutting Plane (click the icon) and Normal → Z (this defines a cut-plane at z = 0
on which the real part of the E-field tangential (to the cut-plane) will be displayed). Click
Plot Type → Animate Fields. (You should vary the z location of the cut-plane to observe
the fields along the length of the filter.)
13
 Question 14: For the microstrip filter, you have plotted the E-field normal to the direction
of propagation in the stopband. Describe what you observe and interpret/explain your obser-
vations.

10. We will now examine at the behaviour of the E-field tangential to the direction of
propagation. In the menu bar at the top of the screen, select Plot Type → Instanta-
neous/Real/Contour/Z, and Sectional View → Fields on Plane (click the icon)/Cutting
Plane (click the icon) and Normal → X (this defines a cut-plane at x = 0 on which the
real part of the E-field in the z-direction will be displayed). Click Plot Type → Animate
Fields.

Question 15: For the microstrip filter, you have plotted the magnitude of the E-field in the
direction of propagation (i.e. the z-direction) in the stopband. Describe what you observe and
interpret/explain your observations.

4 Conclusions

This laboratory has provided an introduction to the use of the numerical simulation package
CST Microwave Studio® . As an example, a microstrip low-pass filter has been implemented
and its performance investigated.

References
[1] Simulia — CST Studio Suite (https://www.3ds.com/products/simulia/cst-studio-suite).

[2] M. J. Neve, “ELECTENG 721/ELECTENG 737 Course Notes: Transmission Lines, Match-
ing, Filters and Standards,” 2025.

[3] Dassault Systems (https://www.3ds.com).

[4] D. B. Davidson, Computational Electromagnetics for RF and Microwave Engineer-

ing, 2nd ed. Cambridge: Cambridge University Press, 2011. [Online]. Available:
http://assets.cambridge.org/97805215/18918/cover/9780521518918.jpg

[5] U. S. Inan and R. A. Marshall, Numerical Electromagnetics: The FDTD

Method. Cambridge: Cambridge University Press, 2011. [Online]. Available:
http://assets.cambridge.org/97805211/90695/cover/9780521190695.jpg

[6] H. A. Wheeler, “Transmission-Line Properties of a Strip on a Dielectric Sheet on a Plane,”

IEEE Trans. Microwave Theory Tech., vol. 25, no. 8, pp. 631–647, Aug. 1977.

[7] E. O. Hammerstad, “Equations for Microstrip Circuit Design,” in Proceedings of 5th Euro-
pean Microwave Conference, 1975, pp. 268–272.

[8] “RT/duroid® 5880 Laminates,” https://www.rogerscorp.com/.

14
A Graphs of Z0 (Ω) vs w

300
h = 3.175 mm
280
h = 1.575 mm
260
h = 0.787 mm

240

220

200

180
Z0 (Ω)

160

140

120

100

80

60

40

20

0
0 1 2 3 4 5

w (mm)

Figure 11: RT/duroid 5880 laminate characteristic impedance Z0 (Ω) vs microstrip line
width w (mm, 0 < w ≤ 5 mm) for various substrate thicknesses h (mm) (substrate
ϵr = 2.2, 35 µm cladding).

15
 80
h = 3.175 mm
h = 1.575 mm
70 h = 0.787 mm

60

50
Z0 (Ω)

40

30

20

10

0
5 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40

w (mm)

Figure 12: RT/duroid 5880 laminate characteristic impedance Z0 (Ω) vs microstrip line
width w (mm, 5 mm < w ≤ 40 mm) for various substrate thicknesses h (mm) (substrate
ϵr = 2.2, 35 µm cladding).

16