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

High Pass Filter

  1. Sep 7, 2010 #1
    Hey everyone I have recently started a circuits class, and am enjoying it. However, the high pass filters have slightly confusing me.

    I recently set up a circuit that was straight from a wall socket meaning that it has a frequency of ~ 60 Hz. My high-pass filter should filter out all of the low frequencies and keep the high ones. My question is, where do these high frequencies originate if the power source only has 60Hz.

    Does it perhaps have something to do with the capacitor discharging? I'm just lost on this one. Any help is greatly appreciated.

    Thanks,
    Dane Peagler
     
  2. jcsd
  3. Sep 7, 2010 #2
    A 60 Hz from wall socket is just 60 Hz. That's all. You can't change it. Sometimes there might be a 120 Hz second harmonic present.

    But if you have a different AC source, like let's say audio from a microphone generating frequencies from 20 Hz to 20 KHz, you could use a filter to filter out the band that you're interested in.
     
  4. Sep 7, 2010 #3

    vk6kro

    User Avatar
    Science Advisor

    A high pass filter on the mains would be unusual. Usually, you want to get rid of the mains rubbish that is not at 60 Hz.

    Safety warning. DO NOT actually connect anything to the mains unless you know exactly what you are doing and can do it safely. High voltages can kill you.

    Every time you operate a switch or turn on a motor, you generate high frequency components that travel along the mains wires.

    Power companies send control pulses along the mains and these can be at hundreds of Hz.

    There have been attempts to send Internet along the mains. These radiate radio interference, so this is a bad idea, but if it existed in your area, you could use a high pass filter to split it off from the 60 Hz high powered sinewave.

    If you especially wanted to listen to all this junk, you could try to filter out the 60 Hz signal and collect the higher frequency stuff with a high pass filter.
     
  5. Sep 7, 2010 #4

    berkeman

    User Avatar

    Staff: Mentor

    I second vk6kro's safety warning. Please do not use the AC Mains as a signal source for experiments unless you have the training and experience to deal with hazardous voltages.

    Common sources of high-frequency noise on AC Mains circuits include dimmable ballasts (to dim incandescent lights), flourescent light ballasts, and phone and TV extender devices that communicate over the powerline.
     
  6. Sep 8, 2010 #5
    Thanks for the concerns everyone. I am doing this experiment in the lab setting with an instructor there, so proper safety is observed.

    But what the lab included was a transformer hooked to the wall socket to step the voltage down to 6.3 V, it then ran into my high pass filter and into the scope for measurement. So in that setup would the transformer be changing the frequency? I don't think it would, but just am curious as to what would be the source of the high frequencies still.
     
  7. Sep 9, 2010 #6

    berkeman

    User Avatar

    Staff: Mentor

    Absolutely the coupling circuit affects the transfer function and what you see downstream. So a big part of your analysis and writeup should be to characterize the transfer function (as a function of frequency) of the coupling circuit. The coupling circuit for the measurent will have HPF and LPF rolloffs that need to be part of the signal analysis end-to-end that you are wanting to do.
     
    Last edited: Sep 9, 2010
  8. Sep 9, 2010 #7

    vk6kro

    User Avatar
    Science Advisor

    The transformer is a good safety feature but it may limit the high frequencies that are available, depending on the frequency response of the transformer.
    Stepping down the mains voltage will also step down the noise and other components above 60 Hz.
    Safety is more important than these concerns, but you should be aware of them.

    Several sources of signal have been suggested, but what you have not told us is the design "knee" frequency of the filter.

    If it was designed for 10 Hz, the 60 Hz signal would pass with only slight attenuation.

    If it was designed for 1000 Hz, depending on the filter, a large amount of the 60 Hz sinewave would be removed.

    If you had trouble getting rid of the 60 Hz sinewave, maybe you would be allowed to try a 60 Hz notch filter which could be better at removing the big 60 Hz signal.

    What are you supposed to be getting from the mains apart from 60 Hz?
     
  9. Sep 9, 2010 #8
    Sorry about leaving out the details. The knee of the filter is >> 60Hz. I'm not sure of the exact value, but I think it is something on the order of 2.7*10^5 Hz.

    All the experiment asks is what is the observed values of frequency. However, for some reason I kept getting ~60Hz and this was just raising some flags, I'm pretty sure the circuit is set up correctly, because the same configuration worked in an earlier section.

    Thanks for all the great information guys.
     
  10. Sep 10, 2010 #9

    vk6kro

    User Avatar
    Science Advisor

    Maybe someone is trying to get you to see AM broadcast signals. These are about 1 MHz +/- 0.5 MHz. These are picked up by the mains power wires.

    However, these are not passed properly through a conventional transformer. The transformer will have capacitance from primary to secondary but this may not produce a voltage across the secondary winding. The whole secondary could have a voltage on it relative to ground.

    You could dispense with the transformer and obtain a long piece of wire, perhaps 50 ft of it and use this as an antenna to pick up such signals.

    What sort of high pass filter are you using? Is it made of inductors and capacitors or does it have an op-amp in it?
     
  11. Sep 10, 2010 #10
    The filter is just a simple RC Circuit with the knee I mentioned earlier.
     
  12. Sep 11, 2010 #11

    vk6kro

    User Avatar
    Science Advisor

    This isn't making much sense.

    If you want to test a filter, you connect a signal generator to the input and an oscilloscope to the output, then vary the frequency and obtain a series of readings of output vs input.

    You then graph the results.

    Maybe after that you can start using it to observe random junk on the mains power supply.

    Simple RC high-pass filters are not very good and they generally give a very gradual rising response.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook