Difference between revisions of "Permittivity and Permeability of Materials Obstacle Course"

From Advanced Projects Lab
Jump to: navigation, search
(Materials to Borrow When Necessary)
(Activities)
 
(One intermediate revision by the same user not shown)
Line 13: Line 13:
 
- Lathe<br>
 
- Lathe<br>
 
- RF Lockin<br>
 
- RF Lockin<br>
- Oscilloscope and Active Probes<br>
+
- Active Probes<br>
- Function Generator<br>
 
  
 
== Activities ==
 
== Activities ==
Line 48: Line 47:
 
<br><br><br><br><br><br><br>
 
<br><br><br><br><br><br><br>
  
=== * Permittivity of a ''Lossy'' Material From a Capacitance Measurement (up to 10 MHz)===
+
=== * Permittivity of a ''Lossy'' Material From a Capacitance Measurement (up to 10 MHz) ===
  
 
[[File:RCVdivider.png|right|800px |thumb]]
 
[[File:RCVdivider.png|right|800px |thumb]]
Line 62: Line 61:
 
<br><br><br><br>
 
<br><br><br><br>
  
Beyond 10MHz lumped circuits become non-ideal (resistors start exhibiting capacitive and inductive effects - analogously with capacitors and inductors - even plain wires). So, to go beyond 10MHz, we need to abandon the lumped capacitance model for determining the permittivity. There are a few techniques for doing this. And the transmission-line technique is one of them.
+
Beyond 10MHz lumped circuits become non-ideal (resistors start exhibiting capacitive and inductive effects - analogously with capacitors and inductors - even plain wires). So, to go beyond 10MHz, we need to consider how the lumped-component model fails for determining the permittivity. We can put the RC voltage divider circuit on a ground-plane surface-mount PCB for higher frequencies or move to the transmission-line method or the waveguide method (next three sections below).
  
=== Transmission Line Techniques (above 50MHz) ===
+
=== * Permittivity of a ''Lossy'' Material From a Capacitance Measurement (up to ~ 500 MHz) ===
 +
small surface mount components on a ground-plane PCB...
 +
 
 +
=== Transmission Line Techniques (above 500 MHz) ===
 +
coming soon...
 +
 
 +
=== Waveguide Techniques (above 500MHz) ===
 
coming soon...
 
coming soon...

Latest revision as of 10:00, 24 May 2018

Permanent Materials

- 6061 3/8" Al rod stock
- Teflon
- Glass microscope slide
- HP Signal Generator (DC-1 GHz)
- Oscilloscope (at least 1GHz bandwidth)
- Miscellaneous electrical components

Materials to Borrow When Necessary

- Milling machine
- Lathe
- RF Lockin
- Active Probes

Activities

Reading

  1. Read the Wikipedia articles on permittivity and permeability. With the help of the instructor or TA try to achieve a physical understanding of just what the permittivity and permeability mean in a bulk material.
  2. Read the first three sections of this paper (pages 1-27). Pay particular attention to the permittivity () / capacitance and permeability () / inductance associations.

Capacitance Techniques (Below 10MHz)

* Permittivity of a Lossless Material From a Capacitance Measurement

Epsiloncap1.png
  1. Place three samples (air, Teflon, glass) between the aligned and polished ends of two 3/8" diameter, 1/2" lengths of 6061 Al rods (as shown at right). should be on the order of 1 mm. Measure the capacitances and, from the known surface area and spacing , determine the material's relative permittivity. (for a capacitor with no fringing fields).
  2. How do your measured permittivity values compare to standard reference values?
  3. Use this web applet to build a capacitor and observe the field lines . Are there fringing fields?

(air): 1.000536
(Teflon): 2.1
(glass): 3.7-10

* A Better Permittivity-Capacitance Measurement of a Lossless Material

Epsiloncap2.png
  1. Use this web applet to build a guarded-electrode capacitor (as shown at the right) and observe the field lines . Are there fringing fields?
  2. Measure the three permittivities (air, Teflon, glass) again using this guarded-electrode setup.
  3. How do these results compare to your first (unguarded) measurements?
  4. How do these results compare to the standard values?




The measurements above for a lossless material amounts to requiring the permittivity to be real (as opposed to complex). However, for a lossy material, the permittivity is complex and we need an additional characteristic (beyond simply the capacitance) to characterize the material. This additional characteristic is the conductance . The measurement below will include the conductance of the material.






* Permittivity of a Lossy Material From a Capacitance Measurement (up to 10 MHz)

RCVdivider.png
  1. Employ the guarded-electrode setup above and measure the lossy material's capacitance and conductance . and , where are, respectively, the real and imaginary parts of the complex permittivity.
    1. Simultaneously measure the voltage and phase across the resistor and capacitor .
    2. The current through the capacitor is given by . The voltage across the capacitor is given by , where is the capacitor's impedance. Solve for the impedance .
    3. The capacitor's admittance is given by , where the (the conductance) and the (the susceptance). Calculate .
    4. Read section 13.1 (pages 106-107) in this [paper].
    5. Calculate the relative permittivity as and .
    6. Do the above procedure for three frequencies at 0.1 MHz, 1MHz and 10 MHz.
    7. Plot and as a function of frequency.





Beyond 10MHz lumped circuits become non-ideal (resistors start exhibiting capacitive and inductive effects - analogously with capacitors and inductors - even plain wires). So, to go beyond 10MHz, we need to consider how the lumped-component model fails for determining the permittivity. We can put the RC voltage divider circuit on a ground-plane surface-mount PCB for higher frequencies or move to the transmission-line method or the waveguide method (next three sections below).

* Permittivity of a Lossy Material From a Capacitance Measurement (up to ~ 500 MHz)

small surface mount components on a ground-plane PCB...

Transmission Line Techniques (above 500 MHz)

coming soon...

Waveguide Techniques (above 500MHz)

coming soon...