Emagtech Wiki - User contributions [en]

RF Tutorial Lesson 4: Analyzing Microstrip Lines & Discontinuities

2016-11-09T21:50:52Z

Shatch:

{{projectinfo|Tutorial| Analyzing Microstrip Lines & Discontinuities |RF46.png|In this project, you will use RF.Spice's line designer and line calculator utilities to build and analyze simple microstrip circuits.|

*Physical Transmission Line
*Microstrip Line
*Device Manager
*Line Designer
*Line Calculator
*Microstrip Discontinuity
|All versions|{{download|http://www.emagtech.com/downloads/ProjectRepo/RFLesson4.zip RF Lesson 4}} }}

== What You Will Learn ==

In this tutorial you will learn how to design and use microstrip lines and components. You will analyze a microstrip double-step and will utilize microstrip discontinuity models to improve the accuracy of your analysis.

== Designing Microstrip Lines ==

[[RF.Spice A/D]] has a large number of physical line calculators and designers. You can access these tools from the Device Manager. Open RF.Spice's Device Manager either from its '''File Menu''' or using the keyboard shortcut {{key|Ctrl+D}}. At the top of Device Manager's '''Tools Menu''', find '''Microstrip Designer''' and open it. For this project, you will use an FR-4 substrate of thickness h = 1.2mm, relative permittivity er = 4.5 and loss tangent tand = 0.02. To design a 50Ω microstrip line, enter these values into the designer dialog and set Z0 = 50. The corresponding width value is computed to be 2.3mm.

<table>
<tr>
<td>
[[File:RF40.png|thumb|left|420px|The RF.Spice A/D Microstrip Designer dialog accessible from the Device Manager.]]
</td>
</tr>
</table>

== Verifying a Simple Microstrip Circuit ==

In order to verify your design, go back to the [[RF.Spice A/D]] Workshop and place a "Microstrip Line" part on the schematic either using '''Menu > Parts > Transmission Lines > Physical Transmission Lines > Microstrip Line''' or using the keyboard shortcut {{key|Alt+T}}. The following is a list of parts needed for this part of the tutorial lesson:

{| border="0"
|-
| valign="top"|
|-
{| class="wikitable"
|-
! scope="col"| Part Name
! scope="col"| Part Type
! scope="col"| Part Value
|-
! scope="row"| AC1
| AC Voltage Source
| 1V
|-
! scope="row"| XMS1
| Microstrip Line
| w = 2.3, h = 1.2, er = 4.45, len = 20, tand = 0.02
|-
! scope="row"| R1
| Resistor
| 50
|-
! scope="row"| R2
| Resistor
| 50
|-
! scope="row"| IN
| Net Marker
| N/A
|}

<table>
<tr>
<td>
[[File:RF44.png|thumb|left|640px|The property dialog of the Microstrip Line device.]]
</td>
</tr>
</table>

Place and connect the parts as shown in the figure below. Ground both negative pins of the microstrip device. Place a Net marker called "IN" (keyboard shortcut: {{key|Alt+N}}) at the input of the microstrip line segment.

<table>
<tr>
<td>
[[File:RF42.png|thumb|left|420px|left|A simple microstrip circuit with a resistive load termination.]]
</td>
</tr>
</table>

Run a Network Analysis of this simple circuit, with Port 1 defined between node IN and the ground. Use the following parameters:

{| border="0"
|-
| valign="top"|
|-
{| class="wikitable"
|-
! scope="row"| Start Frequency
| 100Meg
|-
! scope="row"| Stop Frequency
| 6G
|-
! scope="row"| Steps/Interval
| 100Meg
|-
! scope="row"| Interval Type
| Linear
|-
! scope="row"| Parameter Set
| S
|-
! scope="row"| Graph Type
| Smith
|}

At the end of the simulation, you will see a Smith chart like the one shown below with all the frequency points overlaid at the center of the chart. This means that the input impedance of your circuit at Port 1 is Z<sub>in</sub> = 50Ω at all frequencies. This can happen only if the characteristic impedance of you transmission line segment perfectly matches your termination load.

<table>
<tr>
<td>
[[File:RF43.png|thumb|left|720px|Smith chart for the return loss of the basic resistor-terminated microstrip line segment.]]
</td>
</tr>
</table>

== Calculating Microstrip Line Parameters ==

Next, you will use the Device Manager's "Microstrip Calculator" to find the effective permittivity of your 50Ω microstrip line and its guide wavelength at 2GHz. Open the '''Line Calculator''' from the '''Tools Menu''' of Device Manager. Enter the parameters: w = 2.3mm, h = 1.2mm, er = 4.45, len = 20mm, tand = 0.02, as shown in the figure below.

<table>
<tr>
<td>
[[File:RF41.png|thumb|left|540px|The RF.Spice A/D Microstrip Calculator dialog.]]
</td>
</tr>
</table>

In the other parts of this tutorial lesson and the next one, you will use very wide and very narrow microstrip line segments. Repeat the same calculations for microstrip width values of 0.5mm and 5mm. The results are summarized in the table below:

{| class="wikitable"
|-
! w !! Z<sub>0</sub> !! ε<sub>eff</sub> !! Guide Wavelength λ<sub>g</sub> at 2GHz
|-
| 2.3mm || 49.74Ω || 3.37 || 81.76mm
|-
| 0.5mm || 101.59Ω || 3.06 || 85.79mm
|-
| 5mm || 29.55Ω || 3.61 || 78.91mm
|-
|}

== Analyzing a Microstrip Double Step ==

The characteristic impedance Z0 of a microstrip line is inversely proportional to its width as you saw in the previous section. When you connect two microstrip lines of different widths directly to each other, you create a transmission line discontinuity, which is called an impedance step. In this part of the tutorial lesson, you will analyze a circuit composed of two back-to-back 2.3mm-to-5mm and 5mm-to-2.3mm impedance steps with a 10mm-long, 5mm-wide, microstrip segment in between. You will use 50Ω input and output lines as shown in the opposite figure. A list of the microstrip line segments you need to place in your schematic is given in the table below:

{| class="wikitable"
|-
! Line Segment !! w !! len !! Z<sub>0</sub>
|-
| XMS1 || 2.3mm || 5mm || 50Ω
|-
| XMS2 || 2.3mm || 5mm || 50Ω
|-
| XMS3 || 5mm || 20mm || 30Ω
|-
|}

<table>
<tr>
<td>
[[File:RF45.png|thumb|left|540px|The top view of layout of the microstrip double step structure.]]
</td>
</tr>
</table>

Place the parts and connect them as shown below. Place two "Net" Markers called "IN" and "OUT" at the input and output of your circuit.

<table>
<tr>
<td>
[[File:RF46.png|thumb|left|540px|The schematic of the microstrip double step structure.]]
</td>
</tr>
</table>

Run a Network Analysis of the double step circuit, with Port 1 defined between node IN and the ground and Port 2 defined between node OUT and the ground.

{| border="0"
|-
| valign="top"|
|-
{| class="wikitable"
|-
! scope="row"| Start Frequency
| 100Meg
|-
! scope="row"| Stop Frequency
| 6G
|-
! scope="row"| Steps/Interval
| 10Meg
|-
! scope="row"| Interval Type
| Linear
|-
! scope="row"| Parameter Set
| S
|-
! scope="row"| Graph Type
| Cartesian (Amplitude Only) with Decibels
|}

You should get a graph like the one shown below. Microstrip steps are often used in the design of distributed filters.

<table>
<tr>
<td>
[[File:RF47.png|thumb|left|720px|Graph of the S11 and S21 parameters of the double step circuit over the frequency range 500MHz-6GHz.]]
</td>
</tr>
</table>

== Adding the Microstrip Step Discontinuity Models ==

In the circuit of the previous section, the microstrip line segments of different widths were directly connected to each other, and you ignored the effect of the step discontinuity. However, [[RF.Spice A/D]] also provides a number of microstrip discontinuity models for you to increase the accuracy of your modeling. One of these devices is the two-port "Microstrip Step", which can be accessed from '''Menu > Parts > RF Components > Microstrip Components > Microstrip Step'''. The opposite figure shows the property dialog of this device. It assumes that the first microstrip (w1) at Port 1 is always wider than the second one (w2) at Port 2. Therefore, in order to use this device as a junction from segment L1 to L2, you have to flip it horizontally (keyboard shortcut: {{key|Ctrl+F}}) so that its Port 2 and 1 are connected to L1 and L2, respectively. Use another instance of the same device, but without flipping, as a junction from segment L2 to L3.

The following is a list of parts needed for this part of the tutorial lesson:

{| border="0"
|-
| valign="top"|
|-
{| class="wikitable"
|-
! scope="col"| Part Name
! scope="col"| Part Type
! scope="col"| Part Value
|-
! scope="row"| AC1
| AC Voltage Source
| 1V
|-
! scope="row"| XMS1 - XMS2
| Microstrip Line
| w = 2.3, h = 1.2, er = 4.45, len = 5, tand = 0.02
|-
! scope="row"| XMS3
| Microstrip Line
| w = 5, h = 1.2, er = 4.45, len = 20, tand = 0.02
|-
! scope="row"| X1 - X2
| Microstrip Step
| w1 = 5, w2 = 2.3, h = 1.2, er = 4.45
|-
! scope="row"| IN
| Net Marker
| N/A
|-
! scope="row"| OUT
| Net Marker
| N/A
|}

<table>
<tr>
<td>
[[File:RF51.png|thumb|left|640px|The property dialog of the Microstrip Step device.]]
</td>
</tr>
</table>

Place and connect the parts and flip the microstrip step devices X1 and X2 properly as shown in the figure below:

<table>
<tr>
<td>
[[File:RF48.png|thumb|left|640px|The schematic of the microstrip double step including step discontinuity models.]]
</td>
</tr>
</table>

Run a Network Analysis of your enhanced double step circuit with the following parameters:

{| border="0"
|-
| valign="top"|
|-
{| class="wikitable"
|-
! scope="row"| Start Frequency
| 100Meg
|-
! scope="row"| Stop Frequency
| 6G
|-
! scope="row"| Steps/Interval
| 10Meg
|-
! scope="row"| Interval Type
| Linear
|-
! scope="row"| Parameter Set
| S
|-
! scope="row"| Graph Type
| Cartesian (Amplitude Only) with Decibels
|}

Compare the S-parameters plotted below with the results of the previous case. The insertion loss |S21| is almost identical between the two cases. However, the dip of the return loss |S11| has slightly shifted to the left by about 200MHz after the insertion of the discontinuity models.

<table>
<tr>
<td>
[[File:RF49.png|thumb|left|720px|Graph of S11 and S21 parameters of the double step circuit including step discontinuity models over the frequency range 500MHz-6GHz.]]
</td>
</tr>
</table>
<p> </p>
[[Image:Back_icon.png|40px]] '''[[RF.Spice_A/D#RF.Spice_A.2FD_Tutorials | Back to RF.Spice A/D Tutorial Gateway]]'''

RF Tutorial Lesson 4: Analyzing Microstrip Lines & Discontinuities

2016-11-09T21:24:22Z

Shatch:

Antenna On Platform

2016-10-18T18:19:14Z

Shatch: /* Patch on Hood */

{{projectinfo|Application|Modeling Large Parabolic Reflectors Illuminated By Pyramidal Horn Antennas Using EM.Cube|Hood nearfield.png|In this project, large parabolic reflectors fed by pyramidal horn antennas are modeled and analyzed using EM.Illumina and EM.Tempo.|
*[[EM.Tempo]]
*[[EM.Illumina]]
*Pyramidal Horn
*Parabola
*Object of Revolution
*Field Distribution
*Radiation Pattern
|All versions|None }}
=Patch in Freespace=

[[Image:Patch wizard raw.png|thumb|left|400px|]]

[[Image:Patch variables.png|thumb|left|400px|Default parameterization for the probe-fed patch wizard.]]

[[Image:Patch s11.png|thumb|left|400px|]]

[[Image:Patch rad pattern.png|thumb|left|400px|]]

[[Image:Patch zx cut.png|thumb|left|400px|]]

[[Image:Patch yz cut.png|thumb|left|400px|]]

== Computational Environment ==

The Mirage III CAD model has an approximate length of 15m, a wingspan of 8m, and an approximate height of 4.5m. Expressed in free-space wavelengths at 850 MHz, the approximate dimensions of the aircraft model are 42.5 λ<sub>0</sub> x 22.66 λ<sub>0</sub> x 12.75 λ<sub>0</sub>. Thus, for the purposes of [[EM.Tempo]], we need to solve a region of about 12,279 cubic wavelengths. For problems of this size, a very large CPU memory is needed, and a high-performance, multi-core CPU is desirable to reduce the simulation time.

[https://aws.amazon.com/ Amazon Web Services] allows one to acquire high-performance compute instances on demand, and pay on a per-use basis. To be able to log into an Amazon instance via Remote Desktop Protocol (RDP), the [[EM.Cube]] license must allow terminal services. For the purpose of this project, we used a c4.4xlarge instance running Windows Server 2012. This instance has 30 GB of RAM memory, and 16 virtual CPU cores. The CPU for this instance is an Intel Xeon E5-2666 v3 (Haswell) processor.

<br clear="all"/>

=Patch on Roof=

Materials used in car model:

{| class="wikitable"
|-
! scope="col"| Material
! scope="col"| Relevant Components
of Golf Model
! scope="col"| ε<sub>r</sub>
! scope="col"| σ
|-
| PEC
| car body
| 1
| ∞
|-
| Glass
| car windows
| 6.5
| 0.005
|-
| Plastic
| head-light covers, brake-light covers,
license plate mounts
| 2.2
| 0.0
|-
| Rubber
| tires
| 2.9
| 0.005
|-
| Aluminum
| wheel-rims
| 1
| 3800000
|-
| Cement
| road
| 1.9
| 0.0
|}

Simulation Information:

Mesh size: 220 million cells

Farfield Resolution: 2.5 degrees

Simulation Time: 4 hours, 45 minutes

Typical Performance : 320 MCells/s

Power Threshold: -40 dB

Thread Factor: 8

The thread factor setting essentially tells the FDTD engine how many CPU threads to use during [[EM.Tempo]]'s time-marching loop. For a given system, some experimentation may be needed to determine the best number of threads to use. In many cases, using half of the available hardware concurrency works well. This comes from the fact that many modern processors often have two cores per memory port. In other words, for many problems, the FDTD solver cannot load and store data from CPU memory quickly enough to use all the available threads or hardware concurrency. The extra threads remain idle waiting for the data, and a performance hit is incurred due to the increased thread context switching. [[EM.Cube]] will attempt use a version of the FDTD engine optimized for use with Intel's AVX instruction set, which provides a significant performance boost. If AVX is unavailable, a less optimal version of the engine will be used alternatively.

[[Image:Roof field.png|thumb|left|400px|]]

[[Image:Roof mesh.png|thumb|left|400px|]]

[[Image:Roof mesh settings.png|thumb|left|400px|]]

[[Image:Roof mesh settings advanced.png|thumb|left|400px|]]

By default, [[EM.Tempo]]'s mesher tries to place grid points at the corners of each graphic object's bounding box, and also at any internal boundaries the object may have. For models with a large number of complex objects, this can drive the typical mesh cell size toward the Absolute Minimum Grid Spacing, and result in a much finer mesh than is required. Since the VW Golf model has around 2000 graphic objects, we will turn off these options.

[[Image:Roof patch.png|thumb|left|400px|]]

[[Image:Roof patch mesh.png|thumb|left|400px|]]

[[Image:Roof pattern.png|thumb|left|400px|]]

[[Image:Roof wheel mat.png|thumb|left|400px|]]

[[Image:Roof wheel mat select.png|thumb|left|400px|]]

[[Image:Roof yz cut.png|thumb|left|400px|]]

[[Image:Roof zx cut.png|thumb|left|400px|]]

<br clear="all">

=Patch on Hood=

Simulation Information:

Mesh size: 230 million cells

Farfield Resolution: 2.5 degrees

Simulation Time: 5 hours, 25 minutes

Typical Performance : 320 MCells/s

Power Threshold: -40 dB

Thread Factor: 8

[[Image:Hood mount.png|thumb|left|400px|]]

[[Image:Hood mount mesh detail.png|thumb|left|400px|]]

[[Image:Hood nearfield.png|thumb|left|400px|]]

[[Image:Hood pattern.png|thumb|left|400px|]]

[[Image:Hood yz cut.png|thumb|left|400px|]]

[[Image:Hood zx cut.png|thumb|left|400px|]]

<br clear="all"/>

Antenna On Platform

2016-10-18T18:17:05Z

Shatch: /* Patch on Roof */

{{projectinfo|Application|Modeling Large Parabolic Reflectors Illuminated By Pyramidal Horn Antennas Using EM.Cube|Hood nearfield.png|In this project, large parabolic reflectors fed by pyramidal horn antennas are modeled and analyzed using EM.Illumina and EM.Tempo.|
*[[EM.Tempo]]
*[[EM.Illumina]]
*Pyramidal Horn
*Parabola
*Object of Revolution
*Field Distribution
*Radiation Pattern
|All versions|None }}
=Patch in Freespace=

[[Image:Patch wizard raw.png|thumb|left|400px|]]

[[Image:Patch variables.png|thumb|left|400px|Default parameterization for the probe-fed patch wizard.]]

[[Image:Patch s11.png|thumb|left|400px|]]

[[Image:Patch rad pattern.png|thumb|left|400px|]]

[[Image:Patch zx cut.png|thumb|left|400px|]]

[[Image:Patch yz cut.png|thumb|left|400px|]]

== Computational Environment ==

The Mirage III CAD model has an approximate length of 15m, a wingspan of 8m, and an approximate height of 4.5m. Expressed in free-space wavelengths at 850 MHz, the approximate dimensions of the aircraft model are 42.5 λ<sub>0</sub> x 22.66 λ<sub>0</sub> x 12.75 λ<sub>0</sub>. Thus, for the purposes of [[EM.Tempo]], we need to solve a region of about 12,279 cubic wavelengths. For problems of this size, a very large CPU memory is needed, and a high-performance, multi-core CPU is desirable to reduce the simulation time.

[https://aws.amazon.com/ Amazon Web Services] allows one to acquire high-performance compute instances on demand, and pay on a per-use basis. To be able to log into an Amazon instance via Remote Desktop Protocol (RDP), the [[EM.Cube]] license must allow terminal services. For the purpose of this project, we used a c4.4xlarge instance running Windows Server 2012. This instance has 30 GB of RAM memory, and 16 virtual CPU cores. The CPU for this instance is an Intel Xeon E5-2666 v3 (Haswell) processor.

<br clear="all"/>

=Patch on Roof=

Materials used in car model:

{| class="wikitable"
|-
! scope="col"| Material
! scope="col"| Relevant Components
of Golf Model
! scope="col"| ε<sub>r</sub>
! scope="col"| σ
|-
| PEC
| car body
| 1
| ∞
|-
| Glass
| car windows
| 6.5
| 0.005
|-
| Plastic
| head-light covers, brake-light covers,
license plate mounts
| 2.2
| 0.0
|-
| Rubber
| tires
| 2.9
| 0.005
|-
| Aluminum
| wheel-rims
| 1
| 3800000
|-
| Cement
| road
| 1.9
| 0.0
|}

Simulation Information:

Mesh size: 220 million cells

Farfield Resolution: 2.5 degrees

Simulation Time: 4 hours, 45 minutes

Typical Performance : 320 MCells/s

Power Threshold: -40 dB

Thread Factor: 8

The thread factor setting essentially tells the FDTD engine how many CPU threads to use during [[EM.Tempo]]'s time-marching loop. For a given system, some experimentation may be needed to determine the best number of threads to use. In many cases, using half of the available hardware concurrency works well. This comes from the fact that many modern processors often have two cores per memory port. In other words, for many problems, the FDTD solver cannot load and store data from CPU memory quickly enough to use all the available threads or hardware concurrency. The extra threads remain idle waiting for the data, and a performance hit is incurred due to the increased thread context switching. [[EM.Cube]] will attempt use a version of the FDTD engine optimized for use with Intel's AVX instruction set, which provides a significant performance boost. If AVX is unavailable, a less optimal version of the engine will be used alternatively.

[[Image:Roof field.png|thumb|left|400px|]]

[[Image:Roof mesh.png|thumb|left|400px|]]

[[Image:Roof mesh settings.png|thumb|left|400px|]]

[[Image:Roof mesh settings advanced.png|thumb|left|400px|]]

By default, [[EM.Tempo]]'s mesher tries to place grid points at the corners of each graphic object's bounding box, and also at any internal boundaries the object may have. For models with a large number of complex objects, this can drive the typical mesh cell size toward the Absolute Minimum Grid Spacing, and result in a much finer mesh than is required. Since the VW Golf model has around 2000 graphic objects, we will turn off these options.

[[Image:Roof patch.png|thumb|left|400px|]]

[[Image:Roof patch mesh.png|thumb|left|400px|]]

[[Image:Roof pattern.png|thumb|left|400px|]]

[[Image:Roof wheel mat.png|thumb|left|400px|]]

[[Image:Roof wheel mat select.png|thumb|left|400px|]]

[[Image:Roof yz cut.png|thumb|left|400px|]]

[[Image:Roof zx cut.png|thumb|left|400px|]]

<br clear="all">

=Patch on Hood=

[[Image:Hood mount.png|thumb|left|400px|]]

[[Image:Hood mount mesh detail.png|thumb|left|400px|]]

[[Image:Hood nearfield.png|thumb|left|400px|]]

[[Image:Hood pattern.png|thumb|left|400px|]]

[[Image:Hood yz cut.png|thumb|left|400px|]]

[[Image:Hood zx cut.png|thumb|left|400px|]]

<br clear="all"/>

File:Roof mesh settings advanced.png

2016-10-18T18:11:37Z

Shatch:

Antenna On Platform