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

Problem with series circuit

  1. Aug 30, 2009 #1
    I built a LC series circuit and am getting no results from it. I placed an inductor first, in series with a polarized capacitor, with the negative connection facing towards the inductor.
    What am I doing wrong with this circuit?
     
  2. jcsd
  3. Aug 30, 2009 #2
    There is nothing wrong with connecting an inductor and a capacitor in series. What is it you are trying to achieve?
     
  4. Aug 30, 2009 #3
    I am trying to filter out unwanted frequencies and allow the resonant frequency to pass.
     
  5. Aug 30, 2009 #4
    How did you connect the series LC to the generator? Did you ground the LC series (notch filter) or does the input from the generator hook to the LC and then to load (bandpass filter)?

    Perhaps the Q of the LC is small and you are seeing a broadband response. Also what frequency are trying to filter?
     
  6. Aug 30, 2009 #5
    I am filtering for 60Hz. I connected the generator to the LC in series and had the output run through a multimeter to measure the signal and then to ground.
     
  7. Aug 30, 2009 #6
    What values did you use for L and C?
     
  8. Aug 30, 2009 #7
    4700 micro Farad for c and 1497 micro Henries for L. I matched them up on an online calculator.
     
  9. Aug 30, 2009 #8
    Ok, so "IF" I get this correctly, you need to build something similar to a butterworth lowpass filter. So:

    f = 1/ [2 * pi * sqrt(LC) ] = 60Hz...which is correct.

    Try connect as follows:

    Output of the dirty AC into one end of the inductor, the other end of the inductor into the +side of the cap...and the -vside of the cap into your ground ref.

    Take the clean input from the +side of the cap.

    Make sure you dont screw up the ground refs if you are dealing with multipe ground references. Do this safely...slowly increase the ac source to test this at a low input value.

    let me know how it works out....
     
  10. Aug 30, 2009 #9

    Redbelly98

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Isn't it a bad idea to connect a polarized capacitor to an AC source?
     
  11. Aug 30, 2009 #10
    No big deal for small signals. High capacity polarized caps are used frequently to bypass audio in audio amplifiers.

    LC filter for 60 Hz is not a good idea to do because of large valued and ultra high Q components are required to achieve at least a 10 Hz bandwidth. No wonder you are not observing resonance.

    Instead you should construct an RC type of filter or an active filter which can easily handle the low end of the audio spectrum.
     
  12. Aug 30, 2009 #11

    Redbelly98

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper

    Ah, okay. Thanks for replying.
     
  13. Aug 30, 2009 #12

    vk6kro

    User Avatar
    Science Advisor

    With small signal levels, you probably have another component you may not have considered.
    You would have relatively high impedance drive for this tuned circuit.

    Because these components have very low reactance, you may have something like 50 ohms in series with a 4700 uF capacitor which only has a reactance of 0.56 ohms.
    If you had 10 volts coming out of the generator, the maximum current you could get would be 200 mA. This in a capacitor of 0.56 ohms reactance would give about 0.1 volts.

    The Q of this circuit will be XL/R or 0.56 / 50 or 0.011 so you would probably not see any resonance at all. (Q is a measure of how selective the tuned circuit is.). The bandwidth would be about 500 Hz.

    If you changed your components to these: 13.26 uF and 530 mH, you would be more likely to see resonance effects. In this case, you would have a Q of about 4 and the peak would be visible but still fairly broad. It would have a bandwidth of about 60/4 or 15 Hz ie about +/- 7.5 Hz or 52.5 to 67.5 Hz.



    .
     
  14. Aug 30, 2009 #13
    You didn't state whether this is a signal line or a rectified power voltage. (Rectified power usually has 120 Hz ripple). But the word "resonant" implies that this is a signal line. How much current is in this line? Often the average dc voltage is very near zero volts, in which case I would put little faith in a polarized capacitor. Where does the 60 Hz come from? Is it from lack of shielding or ground currents, or is it in the original signal? Prevention of noise, hum, etc. is often easier than the cure.
     
  15. Aug 31, 2009 #14

    vk6kro

    User Avatar
    Science Advisor

    I am trying to filter out unwanted frequencies and allow the resonant frequency to pass.

    So, he wants to keep the 60 Hz and reject everything else.

    I'd also like to see non-electrolytic capacitors used. The electros are OK for bypassing, but I wouldn't trust them for resonance.

    Band pass op amp circuitry is probably going to be a cheaper option if it is a signal level application. If it is a power aplication, there is no shortage of 60 Hz in the US.
     
  16. Sep 4, 2009 #15
    So to understand this, there are two things wrong with the values for this circuit. First thing is that there is a low reactance between the two components and in series the impedance outside the resonant frequency within a certain degree would be very small. Allowing the bandwidth to be quite large. Secondly the capacitor of 4700uF is too large and not able to function properly with the weak signal from the generator. So the values of 530mH and 13.26uF would work properly for the Q and resonance. My question is, are these common parts and where can I find these electronic components.
     
  17. Sep 4, 2009 #16

    vk6kro

    User Avatar
    Science Advisor

    Yes, that is why you are getting poor resonance.

    The values 530mH and 13.26uF are calculated values. You would not be able to just walk in and buy them. But there is nothing special about them. They just have 200 ohms reactance.

    Do you especially need resonance at 60 Hz or do you just want to see resonance effects?

    I use the following simple program to calculate resonance values:

    https://www.physicsforums.com/attachment.php?attachmentid=18761&d=1241430510

    If you just want to see resonance, you could move to a higher frequency where large reactances are easier to produce and you can use a signal generator. Your coil of 1.497 mH would resonate with a 0.047 uF capacitor at 18974 Hz. That capacitor would be a cheap standard value if you just wanted to try it. but you would need to use an oscilloscope to see the output properly.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Problem with series circuit
  1. Series RLC Circuit (Replies: 5)

Loading...