Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Question on tank circuit

  1. Jun 23, 2009 #1
    I have built a parallel LC resonator to tune into a specific frequency. the frequency I am tuning into is 60Hz. The problem I am having is that I matched the value of the inductor and the value of the capacitor to match at that frequency within certain tolerance, And the is no signal being filtered. The frequencies are passing right through to the ground connection. Does anyone know I am doing wrong?
     
  2. jcsd
  3. Jun 23, 2009 #2

    negitron

    User Avatar
    Science Advisor

    There's any number of possible things wrong, but without a schematic I can't even begin to speculate. Including details about what you expect the circuit to accomplish specifically and details about the components used (for example, are the caps aluminum electrolytics, mylar, poly or are the inductors air core, ferrites, tape wound, laminated and which sort of material; silicon steel, 49 material ferrite, etc.) will also be helpful.
     
  4. Jun 23, 2009 #3
    What I'm trying to do is to separate 60Hz from a wide range of frequencies. the capacitor is a non-polar made from eight polar electrolytic caps. arranged in parallel. The inductor is a iron core toroidal inductor. Also according to the calculations the impedance of the circuit is 0.5 ohm far less that the impedance of the source. If that helps.
     
  5. Jun 23, 2009 #4

    berkeman

    User Avatar

    Staff: Mentor

    You don't make a non-polar capacitor by putting polar capacitors in parallel. You can make a non-polar capacitor of half value by putting two polar capacitors in series, with the "+" ends together, for example. This is done in audio circuits frequently.
     
  6. Jun 25, 2009 #5

    vk6kro

    User Avatar
    Science Advisor

    Try recalculating with a reactance of about 200 ohms. O.53 H and 13.26 uF preferably not electrolytic.

    Your circuit is basically a short circuit anyway at 60 Hz. The signal source would not be able to deliver the sort of circulating current needed to develop resonance voltages across such low impedance components.

    A much better way is to make a high pass filter for 100 Hz rolloff to get rid of 60 hz , or to make an active band pass filter for 60 Hz if you want the 60 Hz.
     
    Last edited: Jun 25, 2009
  7. Jun 25, 2009 #6
    Here are two tank filters suggested above, one a series resonant and the other parallel. with the frequency swept from 40 Hz to 80 Hz. See attached thumbnails. In this case, L=0.53 Henrys, and C = 13.2 microFarads. The impedance is sqrt(L/C) = 200 ohms.
     

    Attached Files:

    Last edited: Jun 25, 2009
  8. Jun 25, 2009 #7

    vk6kro

    User Avatar
    Science Advisor

    Great graphs, Bob S.
    Was that done with Spice?

    I guess 1 ohm resistance is a bit difficult for a 0.53 H inductance but if it could be held below 20 ohms, to get a Q of at least 10, you would still get reasonable selectivity.
     
  9. Jun 25, 2009 #8
    The circuit modeling and frequency sweeps were done with LTSpice IV, free from Linear Technology Corp.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Question on tank circuit
  1. Tank circuit (Replies: 10)

  2. Tank Circuit Problem (Replies: 25)

Loading...