Difference between revisions of "EM.Tempo Tutorial Lesson 11: A Time-Domain Study of Wave Propagation in Anisotropic Materials"
Kazem Sabet (Talk | contribs) (Created page with "{{projectinfo|Tutorial|A Time-Domain Study of Wave Propagation in Anisotropic Materials|FDTD759.png|In this project, you will investigate wave propagation in isotropic and ani...") |
Kazem Sabet (Talk | contribs) |
||
Line 26: | Line 26: | ||
#Center Frequency: 1GHz | #Center Frequency: 1GHz | ||
#Bandwidth: 2GHz | #Bandwidth: 2GHz | ||
− | |||
In this tutorial lesson, you will explore time-domain propagation of electromagnetic signals first in (isotropic) dielectric media and then in anisotropic media. | In this tutorial lesson, you will explore time-domain propagation of electromagnetic signals first in (isotropic) dielectric media and then in anisotropic media. | ||
==Modeling a Homogeneous Dielectric Medium== | ==Modeling a Homogeneous Dielectric Medium== | ||
− | |||
Define a dielectric material group called Dielectric_1 in the Navigation Tree. Call the material DIEL and set its material properties as ε<sub>r</sub> = 9, μ<sub>r</sub> = 1, σ = 0, and σ<sub>m</sub> = 0. Under this material group, draw a large dielectric box of dimensions 1200mm × 600mm × 600mm. | Define a dielectric material group called Dielectric_1 in the Navigation Tree. Call the material DIEL and set its material properties as ε<sub>r</sub> = 9, μ<sub>r</sub> = 1, σ = 0, and σ<sub>m</sub> = 0. Under this material group, draw a large dielectric box of dimensions 1200mm × 600mm × 600mm. | ||
Define a Z-directed "Ideal Source" called IS_1 with default settings and place it at (-300mm, 0, 0). Define two Z-directed field probes called Probe_1 and Probe_2, and place them at (-300mm, 0, 0) and (300mm, 0, 0), respectively. The first probe is collocated with the ideal source. | Define a Z-directed "Ideal Source" called IS_1 with default settings and place it at (-300mm, 0, 0). Define two Z-directed field probes called Probe_1 and Probe_2, and place them at (-300mm, 0, 0) and (300mm, 0, 0), respectively. The first probe is collocated with the ideal source. | ||
− | |||
{{Note| In [[EM.Cube]]'s [[FDTD Module]], an ideal source is a voltage source placed between two adjacent mesh nodes, and acts like a short dipole source in [[EM.Cube]]'s other modules.}} | {{Note| In [[EM.Cube]]'s [[FDTD Module]], an ideal source is a voltage source placed between two adjacent mesh nodes, and acts like a short dipole source in [[EM.Cube]]'s other modules.}} | ||
− | |||
Open the Domain Settings dialog and set all the six offset [[parameters]] in ±X, ±Y, and ±Z equal to zero. The boundary walls at the six sides of the computational domain box are assumed to be perfectly matched layers (PML) by default. This means that they absorb incident waves and prevent them from reflecting back into the domain. By settings all the offsets equal to zero, your dielectric box effectively extends to infinity in all the six principal directions. In other words, your structure now looks like a homogenous dielectric medium. | Open the Domain Settings dialog and set all the six offset [[parameters]] in ±X, ±Y, and ±Z equal to zero. The boundary walls at the six sides of the computational domain box are assumed to be perfectly matched layers (PML) by default. This means that they absorb incident waves and prevent them from reflecting back into the domain. By settings all the offsets equal to zero, your dielectric box effectively extends to infinity in all the six principal directions. In other words, your structure now looks like a homogenous dielectric medium. | ||
− | |||
The following table summarizes all the object definitions for your project: | The following table summarizes all the object definitions for your project: | ||
Line 80: | Line 75: | ||
| (300mm, 0, 0) | | (300mm, 0, 0) | ||
|} | |} | ||
− | |||
<table> | <table> | ||
Line 92: | Line 86: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
Set the FDTD mesh density to 12 cells per wavelength and keep all the other default mesh settings. Open the FDTD Engine Settings dialog from inside the Run dialog and set the Termination Criterion by specifying the "End Time" to 3,000 time steps. Run an FDTD analysis. At the end of the simulation, open the Data Manager and plot the time-domain electric field data for Probe_1 and Probe_2, as shown in the figures below: | Set the FDTD mesh density to 12 cells per wavelength and keep all the other default mesh settings. Open the FDTD Engine Settings dialog from inside the Run dialog and set the Termination Criterion by specifying the "End Time" to 3,000 time steps. Run an FDTD analysis. At the end of the simulation, open the Data Manager and plot the time-domain electric field data for Probe_1 and Probe_2, as shown in the figures below: | ||
− | |||
<table> | <table> | ||
Line 107: | Line 99: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
Probe_1 shows the ideal source waveform, which is a very compact modulated Gaussian waveform. Note that the center frequency of this waveform is 1GHz and its bandwidth is 2GHz. As you can see from the above figure, the source waveform has a delay of 2.173ns. This is the delay for an optimal modulated Gaussian waveform calculated from the following equation: | Probe_1 shows the ideal source waveform, which is a very compact modulated Gaussian waveform. Note that the center frequency of this waveform is 1GHz and its bandwidth is 2GHz. As you can see from the above figure, the source waveform has a delay of 2.173ns. This is the delay for an optimal modulated Gaussian waveform calculated from the following equation: | ||
− | |||
<math> T = \frac{(4.5)(0.966)}{BW} = \frac{(4.5)(0.966)}{2e+9} = 2.173 \times 10^{-9} s </math> | <math> T = \frac{(4.5)(0.966)}{BW} = \frac{(4.5)(0.966)}{2e+9} = 2.173 \times 10^{-9} s </math> | ||
− | |||
On the other hand, the phase velocity in the dielectric medium is v<sub>p</sub> = c/√ε<sub>r</sub> = c/3 = 1e+8 m/s. The distance between the source and observation probe is 600mm. The time it takes the source signal to reach the probe is therefore Δt = (0.6m)/v<sub>p</sub> = (0.6m)/(1e+8m/s) = 6e-9s = 6ns. With the initial 2.173ns delay, the peak signal must reach Probe_2 at t = 8.173ns, which is clearly seen from the above figure. | On the other hand, the phase velocity in the dielectric medium is v<sub>p</sub> = c/√ε<sub>r</sub> = c/3 = 1e+8 m/s. The distance between the source and observation probe is 600mm. The time it takes the source signal to reach the probe is therefore Δt = (0.6m)/v<sub>p</sub> = (0.6m)/(1e+8m/s) = 6e-9s = 6ns. With the initial 2.173ns delay, the peak signal must reach Probe_2 at t = 8.173ns, which is clearly seen from the above figure. | ||
− | |||
Next, change the relative permittivity of the Dielectric_1 group to ε<sub>r</sub> = 4 and run the same simulation once again. The figure below shows the new waveform at Probe_2. in this case, the phase velocity in the dielectric medium is v<sub>p</sub> = c/√ε<sub>r</sub> = c/2 = 1.5e+8 m/s. The time it takes the source signal to reach the probe is therefore Δt = (0.6m)/v<sub>p</sub> = (0.6m)/(1.5e+8m/s) = 4e-9s = 4ns. With the initial 2.173ns delay, the peak signal reaches Probe_2 at t = 6.173ns, which is 2ns earlier than the previous case. | Next, change the relative permittivity of the Dielectric_1 group to ε<sub>r</sub> = 4 and run the same simulation once again. The figure below shows the new waveform at Probe_2. in this case, the phase velocity in the dielectric medium is v<sub>p</sub> = c/√ε<sub>r</sub> = c/2 = 1.5e+8 m/s. The time it takes the source signal to reach the probe is therefore Δt = (0.6m)/v<sub>p</sub> = (0.6m)/(1.5e+8m/s) = 4e-9s = 4ns. With the initial 2.173ns delay, the peak signal reaches Probe_2 at t = 6.173ns, which is 2ns earlier than the previous case. | ||
− | |||
<table> | <table> | ||
Line 133: | Line 120: | ||
In this part of the tutorial lesson, you will consider a dielectric slab of finite thickness but with infinite lateral extents. To accomplish this, all you need to do is to keep the zero domain offsets in the ±X and ±Y directions and move the PML boundary walls away from the dielectric box at the top and bottom of the computational domain by a quarter free-space wavelength. For this purpose, set the values of the ±Z offsets equal to 0.25. The opposite figure shows the geometry setup for the dielectric slab. | In this part of the tutorial lesson, you will consider a dielectric slab of finite thickness but with infinite lateral extents. To accomplish this, all you need to do is to keep the zero domain offsets in the ±X and ±Y directions and move the PML boundary walls away from the dielectric box at the top and bottom of the computational domain by a quarter free-space wavelength. For this purpose, set the values of the ±Z offsets equal to 0.25. The opposite figure shows the geometry setup for the dielectric slab. | ||
− | |||
In the boundless homogeneous medium of the previous part, you could assume a direct ray carrying the source signal from the source location to the probe location 600mm away. In the case of the dielectric slab, besides the direct ray, there will be a number of reflected rays as shown in the figure below. These rays hit the top or bottom air-dielectric interfaces and are reflected towards the probe. It is evident that the optical paths the reflected rays traverse to reach the observation point at the location of Probe_2 are longer than the optical path of the direct ray, which is 600mm. | In the boundless homogeneous medium of the previous part, you could assume a direct ray carrying the source signal from the source location to the probe location 600mm away. In the case of the dielectric slab, besides the direct ray, there will be a number of reflected rays as shown in the figure below. These rays hit the top or bottom air-dielectric interfaces and are reflected towards the probe. It is evident that the optical paths the reflected rays traverse to reach the observation point at the location of Probe_2 are longer than the optical path of the direct ray, which is 600mm. | ||
− | |||
<table> | <table> | ||
Line 145: | Line 130: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
The following table shows the optical path lengths and travel times for the three rays shown in the above figure. Note that a phase velocity of v<sub>p</sub> = c/3 = 1e+8 m/s has been used for the calculations. | The following table shows the optical path lengths and travel times for the three rays shown in the above figure. Note that a phase velocity of v<sub>p</sub> = c/3 = 1e+8 m/s has been used for the calculations. | ||
Line 177: | Line 161: | ||
| 15.593ns | | 15.593ns | ||
|} | |} | ||
− | |||
Set the permittivity of the Dielectric_1 group back to ε<sub>r</sub> = 9, and run an FDTD analysis of your dielectric slab with the same ideal source and field probes of the previous part. In the figure below, you can clearly see three distinct modulated Gaussian pluses corresponding to Rays 1, 2, and 3, with their peaks centered at times 8.173ns, 10.658ns and 15.593ns, respectively. | Set the permittivity of the Dielectric_1 group back to ε<sub>r</sub> = 9, and run an FDTD analysis of your dielectric slab with the same ideal source and field probes of the previous part. In the figure below, you can clearly see three distinct modulated Gaussian pluses corresponding to Rays 1, 2, and 3, with their peaks centered at times 8.173ns, 10.658ns and 15.593ns, respectively. | ||
− | |||
<table> | <table> | ||
Line 189: | Line 171: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
Next, change the permittivity of the Dielectric_1 group to ε<sub>r</sub> = 4, and run another FDTD analysis. In this case, the phase velocity is v<sub>p</sub> = c/2 = 1.5e+8 m/s. The following table shows the optical path lengths and travel times for the three rays: | Next, change the permittivity of the Dielectric_1 group to ε<sub>r</sub> = 4, and run another FDTD analysis. In this case, the phase velocity is v<sub>p</sub> = c/2 = 1.5e+8 m/s. The following table shows the optical path lengths and travel times for the three rays: | ||
Line 221: | Line 202: | ||
| 11.18ns | | 11.18ns | ||
|} | |} | ||
− | |||
You are going to run two simulations at this point. In the first simulation, keep the Z-directed ideal source and Z-directed field probes as in the earlier parts. In the second simulation, change the direction of the ideal source and field probes to +Y. Run an FDTD analysis for each case and compare the output waveforms at Probe_2. Both of the figures below show three distinct modulated Gaussian pluses corresponding to Rays 1, 2, and 3, with their peaks centered at times 6.173ns, 7.83ns and 11.18ns, respectively. Note that in the case of Z-directed and Y-directed sources, the reflected rays have parallel and perpendicular polarizations with respect to the plane of incidence, respectively. The polarization type explains the inversion of the fields of the reflected ray in the first case and no field inversion in the second case. Also,, note that the doubly reflected Ray 3 undergoes two field inversions in the first case. | You are going to run two simulations at this point. In the first simulation, keep the Z-directed ideal source and Z-directed field probes as in the earlier parts. In the second simulation, change the direction of the ideal source and field probes to +Y. Run an FDTD analysis for each case and compare the output waveforms at Probe_2. Both of the figures below show three distinct modulated Gaussian pluses corresponding to Rays 1, 2, and 3, with their peaks centered at times 6.173ns, 7.83ns and 11.18ns, respectively. Note that in the case of Z-directed and Y-directed sources, the reflected rays have parallel and perpendicular polarizations with respect to the plane of incidence, respectively. The polarization type explains the inversion of the fields of the reflected ray in the first case and no field inversion in the second case. Also,, note that the doubly reflected Ray 3 undergoes two field inversions in the first case. | ||
Line 236: | Line 216: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
==Modeling an Anisotropic Material Medium== | ==Modeling an Anisotropic Material Medium== | ||
Line 285: | Line 264: | ||
| (300mm, 0, 0) | | (300mm, 0, 0) | ||
|} | |} | ||
− | |||
Set all the six domain offset values to zero in all directions. Your structure now models an anisotropic medium of infinite extents. | Set all the six domain offset values to zero in all directions. Your structure now models an anisotropic medium of infinite extents. | ||
{{Note| In the current version of [[EM.Cube]]'s [[FDTD Module]], anisotropic materials with nonzero off-diagonal constitutive tensor elements must be finite-sized and cannot touch the PML boundary wall. You can model infinite uniaxial anisotropic media with nonzero diagonal elements only.}} | {{Note| In the current version of [[EM.Cube]]'s [[FDTD Module]], anisotropic materials with nonzero off-diagonal constitutive tensor elements must be finite-sized and cannot touch the PML boundary wall. You can model infinite uniaxial anisotropic media with nonzero diagonal elements only.}} | ||
− | |||
<table> | <table> | ||
Line 302: | Line 279: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
Run an FDTD analysis of your anisotropic medium. In this case, there will be two direct rays from the two ideal sources to the observation point. One ray is Z-polarized and sees a medium relative permittivity of 9, while the other ray is Y-polarized and sees a medium relative permittivity of 4. Note that the phase velocities of the two rays are different and equal to v<sub>pz</sub> = c/3 = 1e+8m/s and v<sub>py</sub> = c/2 = 1.5e+8m/s, respectively. The figures below show the Z and Y components of the electric field at the observation point where Probe_2 and Probe_3 are located. Also, note that similarity of these waveforms to those of the homogeneous dielectric medium with permittivity values of 9 and 4 in the first part of this tutorial lesson. | Run an FDTD analysis of your anisotropic medium. In this case, there will be two direct rays from the two ideal sources to the observation point. One ray is Z-polarized and sees a medium relative permittivity of 9, while the other ray is Y-polarized and sees a medium relative permittivity of 4. Note that the phase velocities of the two rays are different and equal to v<sub>pz</sub> = c/3 = 1e+8m/s and v<sub>py</sub> = c/2 = 1.5e+8m/s, respectively. The figures below show the Z and Y components of the electric field at the observation point where Probe_2 and Probe_3 are located. Also, note that similarity of these waveforms to those of the homogeneous dielectric medium with permittivity values of 9 and 4 in the first part of this tutorial lesson. | ||
− | |||
<table> | <table> | ||
Line 317: | Line 292: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
[[Image:FDTD764.png|thumb|450px|The geometry of the laterally infinite anisotropic slab with zero domain offsets at the four side boundary walls.]] | [[Image:FDTD764.png|thumb|450px|The geometry of the laterally infinite anisotropic slab with zero domain offsets at the four side boundary walls.]] | ||
Line 324: | Line 298: | ||
Take the geometry of the anisotropic structure from the last part, keep the zero values for the ±X and ±Y domain offsets, and change the value of the ±Z domain offsets to 0.25 as in the case of the dielectric slab. Also keep the two collocated ideal sources IS_1 and IS_2 and the two collocated field probes Probe_2 and Porbe_3 from the last part. Run an FDTD analysis of your anisotropic slab and plot the field probe data for the Z and Y components of the electric field. | Take the geometry of the anisotropic structure from the last part, keep the zero values for the ±X and ±Y domain offsets, and change the value of the ±Z domain offsets to 0.25 as in the case of the dielectric slab. Also keep the two collocated ideal sources IS_1 and IS_2 and the two collocated field probes Probe_2 and Porbe_3 from the last part. Run an FDTD analysis of your anisotropic slab and plot the field probe data for the Z and Y components of the electric field. | ||
− | |||
<table> | <table> | ||
Line 336: | Line 309: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
From the above figures, you can see that the first modulated Gaussian pulse of the Ez component (Probe_2) arrives at 8.173ns, while the first modulated Gaussian pulse of the Ey component (Probe_3) arrives at 6.173ns. These are consistent with the results you got earlier for the isotropic dielectric slab. The first Ez pulse resembles that of an isotropic dielectric slab with ε<sub>r</sub> = 9 excited by a Z-directed source. The first Ey pulse resembles that of an isotropic dielectric slab with ε<sub>r</sub> = 4 excited by a Y-directed source. However, the pulses corresponding to the reflected rays are completely different than their isotropic counterparts. This is due to the anisotropic nature of your new slab. As shown in the figure below, the reflected rays are decomposed to parallel and perpendicular polarizations. The parallel-polarized reflected rays have both Z- and X-directed field components. Since ε<sub>rxx</sub> = 2, the X-polarized reflected rays will experience a faster phase velocity of v<sub>px</sub> = c/√2 = 2.12e+8m/s and arrive at the observation point earlier than the Z-polarized reflected rays. | From the above figures, you can see that the first modulated Gaussian pulse of the Ez component (Probe_2) arrives at 8.173ns, while the first modulated Gaussian pulse of the Ey component (Probe_3) arrives at 6.173ns. These are consistent with the results you got earlier for the isotropic dielectric slab. The first Ez pulse resembles that of an isotropic dielectric slab with ε<sub>r</sub> = 9 excited by a Z-directed source. The first Ey pulse resembles that of an isotropic dielectric slab with ε<sub>r</sub> = 4 excited by a Y-directed source. However, the pulses corresponding to the reflected rays are completely different than their isotropic counterparts. This is due to the anisotropic nature of your new slab. As shown in the figure below, the reflected rays are decomposed to parallel and perpendicular polarizations. The parallel-polarized reflected rays have both Z- and X-directed field components. Since ε<sub>rxx</sub> = 2, the X-polarized reflected rays will experience a faster phase velocity of v<sub>px</sub> = c/√2 = 2.12e+8m/s and arrive at the observation point earlier than the Z-polarized reflected rays. | ||
− | |||
<table> | <table> | ||
Line 348: | Line 319: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
The anisotropic material just considered is typical of a biaxial optical crystal, where all the three diagonal permittivity elements have different values. Next, open the property dialog of the ANI_1 material group and change the diagonal permittivity elements to ε<sub>rx</sub> = 9, ε<sub>ry</sub> = 4, and ε<sub>rz</sub> = 9. This material is now typical of a uniaxial optical crystal, where two of the diagonal permittivity elements have identical values different than the third element. | The anisotropic material just considered is typical of a biaxial optical crystal, where all the three diagonal permittivity elements have different values. Next, open the property dialog of the ANI_1 material group and change the diagonal permittivity elements to ε<sub>rx</sub> = 9, ε<sub>ry</sub> = 4, and ε<sub>rz</sub> = 9. This material is now typical of a uniaxial optical crystal, where two of the diagonal permittivity elements have identical values different than the third element. | ||
Run an FDTD analysis of your new anisotropic slab and plot the field probe data for the Z and Y components of the electric field. | Run an FDTD analysis of your new anisotropic slab and plot the field probe data for the Z and Y components of the electric field. | ||
− | |||
<table> | <table> | ||
Line 365: | Line 334: | ||
</tr> | </tr> | ||
</table> | </table> | ||
− | |||
In this case, both the Z- and X-polarized reflected rays with parallel polarization see the same relative permittivity of 9 and experience the same phase velocity of v<sub>p||</sub> = c/3 = 1e+8m/s. On the other hand, the Y-polarized reflected rays with perpendicular polarization see a relative permittivity of 4 and experience a phase velocity of v<sub>py</sub> = c/2 = 1.5e+8m/s. As a result, the second and third Ez pulses resemble those of an isotropic dielectric slab with ε<sub>r</sub> = 9 excited by a Z-directed source. The second and third Ey pulses resemble those of an isotropic dielectric slab with ε<sub>r</sub> = 4 excited by a Y-directed source. | In this case, both the Z- and X-polarized reflected rays with parallel polarization see the same relative permittivity of 9 and experience the same phase velocity of v<sub>p||</sub> = c/3 = 1e+8m/s. On the other hand, the Y-polarized reflected rays with perpendicular polarization see a relative permittivity of 4 and experience a phase velocity of v<sub>py</sub> = c/2 = 1.5e+8m/s. As a result, the second and third Ez pulses resemble those of an isotropic dielectric slab with ε<sub>r</sub> = 9 excited by a Z-directed source. The second and third Ey pulses resemble those of an isotropic dielectric slab with ε<sub>r</sub> = 4 excited by a Y-directed source. | ||
+ | <p> </p> | ||
+ | [[Image:Back_icon.png|40px]] '''[[EM.Cube#EM.Tempo_Tutorial_Lessons | Back to EM.Tempo Tutorial Gateway]]''' | ||
{{FDTD Details}} | {{FDTD Details}} |
Latest revision as of 13:10, 8 September 2015
Contents
Objective:
To build large unbounded and finite-sized material media featuring dielectric and anisotropic materials and study the time-domain wave propagation in such media.
What You Will Learn:
In this tutorial you will learn how to use computational domain offset parameters to build unbounded and finite-sized material media. You will work with anisotropic material types and will also define simple ideal sources.
Getting Started
Open the EM.Cube application and switch to FDTD Module. Start a new project with the following attributes:
- Name: FDTDLesson16
- Length Units: mm
- Frequency Units: GHz
- Center Frequency: 1GHz
- Bandwidth: 2GHz
In this tutorial lesson, you will explore time-domain propagation of electromagnetic signals first in (isotropic) dielectric media and then in anisotropic media.
Modeling a Homogeneous Dielectric Medium
Define a dielectric material group called Dielectric_1 in the Navigation Tree. Call the material DIEL and set its material properties as εr = 9, μr = 1, σ = 0, and σm = 0. Under this material group, draw a large dielectric box of dimensions 1200mm × 600mm × 600mm.
Define a Z-directed "Ideal Source" called IS_1 with default settings and place it at (-300mm, 0, 0). Define two Z-directed field probes called Probe_1 and Probe_2, and place them at (-300mm, 0, 0) and (300mm, 0, 0), respectively. The first probe is collocated with the ideal source.
In EM.Cube's FDTD Module, an ideal source is a voltage source placed between two adjacent mesh nodes, and acts like a short dipole source in EM.Cube's other modules. |
Open the Domain Settings dialog and set all the six offset parameters in ±X, ±Y, and ±Z equal to zero. The boundary walls at the six sides of the computational domain box are assumed to be perfectly matched layers (PML) by default. This means that they absorb incident waves and prevent them from reflecting back into the domain. By settings all the offsets equal to zero, your dielectric box effectively extends to infinity in all the six principal directions. In other words, your structure now looks like a homogenous dielectric medium.
The following table summarizes all the object definitions for your project:
Set the FDTD mesh density to 12 cells per wavelength and keep all the other default mesh settings. Open the FDTD Engine Settings dialog from inside the Run dialog and set the Termination Criterion by specifying the "End Time" to 3,000 time steps. Run an FDTD analysis. At the end of the simulation, open the Data Manager and plot the time-domain electric field data for Probe_1 and Probe_2, as shown in the figures below: Probe_1 shows the ideal source waveform, which is a very compact modulated Gaussian waveform. Note that the center frequency of this waveform is 1GHz and its bandwidth is 2GHz. As you can see from the above figure, the source waveform has a delay of 2.173ns. This is the delay for an optimal modulated Gaussian waveform calculated from the following equation: [math] T = \frac{(4.5)(0.966)}{BW} = \frac{(4.5)(0.966)}{2e+9} = 2.173 \times 10^{-9} s [/math] On the other hand, the phase velocity in the dielectric medium is vp = c/√εr = c/3 = 1e+8 m/s. The distance between the source and observation probe is 600mm. The time it takes the source signal to reach the probe is therefore Δt = (0.6m)/vp = (0.6m)/(1e+8m/s) = 6e-9s = 6ns. With the initial 2.173ns delay, the peak signal must reach Probe_2 at t = 8.173ns, which is clearly seen from the above figure. Next, change the relative permittivity of the Dielectric_1 group to εr = 4 and run the same simulation once again. The figure below shows the new waveform at Probe_2. in this case, the phase velocity in the dielectric medium is vp = c/√εr = c/2 = 1.5e+8 m/s. The time it takes the source signal to reach the probe is therefore Δt = (0.6m)/vp = (0.6m)/(1.5e+8m/s) = 4e-9s = 4ns. With the initial 2.173ns delay, the peak signal reaches Probe_2 at t = 6.173ns, which is 2ns earlier than the previous case. Modeling a Dielectric SlabIn this part of the tutorial lesson, you will consider a dielectric slab of finite thickness but with infinite lateral extents. To accomplish this, all you need to do is to keep the zero domain offsets in the ±X and ±Y directions and move the PML boundary walls away from the dielectric box at the top and bottom of the computational domain by a quarter free-space wavelength. For this purpose, set the values of the ±Z offsets equal to 0.25. The opposite figure shows the geometry setup for the dielectric slab. In the boundless homogeneous medium of the previous part, you could assume a direct ray carrying the source signal from the source location to the probe location 600mm away. In the case of the dielectric slab, besides the direct ray, there will be a number of reflected rays as shown in the figure below. These rays hit the top or bottom air-dielectric interfaces and are reflected towards the probe. It is evident that the optical paths the reflected rays traverse to reach the observation point at the location of Probe_2 are longer than the optical path of the direct ray, which is 600mm. The following table shows the optical path lengths and travel times for the three rays shown in the above figure. Note that a phase velocity of vp = c/3 = 1e+8 m/s has been used for the calculations.
|