Help with filters

Discussion in 'General Electronics Chat' started by c0de3, Apr 8, 2012.

  1. c0de3

    Thread Starter Active Member

    May 1, 2009
    50
    1
    Hi all! I’ve been working through some basic lessons in electronics. I’m working on the capacitor section that deals with using capacitors and resistors as a filter. So I set a simple test circuit. I’ve setup my computer to create a sine wave via the sounds card. I then connect a wire from the sound card to my test board. On the test board I have a .22uf capacitor and a 1.5 ohm resistor (+/- 20%). The configuration is attached. Then I connect to my scope. I have tried to setup the capacitor and resistor as both a high and low pass filter. However, what I’m seeing is that in low pass, I’m blocking all (at least all that my 20Mhz scope can see) AC, and in High Pass I’m passing all AC.

    My understanding is that with a 1.5Ohm resistor and a .22uf cap I should be dealing with ~3khz. Meaning the low pass should pass anything below that, and the high pass should pass anything above that. However, that isn’t what I see… Is my math wrong? Where am I going wrong?


    Attached is a basic drawing of my low pass configuration. Thanks for any tips and help!
     
  2. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    In the low pass configuration shown, the input does not have a resistance. Therefore the reactive component of the capacitor has nothing to work against.

    I'm sure there is a resistance in the sound card, but it isn't what you expect.

    When you flip this around as a high pass filter, the sound card can not supply enough current to drive 1.5 ohms as a load.
     
  3. c0de3

    Thread Starter Active Member

    May 1, 2009
    50
    1
    Thanks for the help! So I need more than the 1.5 ohm resistor? Is there a rule of thumb for what size you need? When you say "input" do you mean the sine wave coming from the PC? Sorry for simple questions, I just to make sure i understand.

    Just to be clear I do get a sine wave on my scope. It's just it either always gets passed, or always gets blocked. I'm expecting to see it blocked only if it is above (or below depending on high or low pass configuation) the frequency from my formula. (f=CR) (3030HZ = .22uf * 1.5 ohm) But if I set the sine wave to 4Khz or 2Khz they are both acting the same... instead of one passing and one being blocked.


    Again, thanks for help.
     
  4. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    Yes, the sine wave is what you are "inputting" to your filter. A sound card expects to see speakers, not 1.5 ohm resistor. Scale your components up to maybe 1000 ohms so the sound card doesn't suffer from lack of ability to supply current and the resistance of the sound card becomes insignificant compared to the input resistance of the filter.

    and remember that the low pass input resistance is calculated with the capacitor, not the output resistance.

    "Rule of thumb" The impedance of the filter must be at least 10X the impedance of the source in order to not load down the source of the sinewave.
     
  5. c0de3

    Thread Starter Active Member

    May 1, 2009
    50
    1
    Thanks again for the help #12. I understand my problem now. I will increase the resistance. The issue is now I'm going to need a super low value capacitor as the simple sound card software only goes down to 100hz. I think I'd need a .01uF cap to get to 100hz with a 1k ohm resistor.


    Can you give me an example of this, I'm not 100% sure what you mean here?
     
  6. #12

    Expert

    Nov 30, 2010
    16,257
    6,757
    Look at the filter as a voltage divider. The input resistance is in series with the capacitor (to ground). If the input resistance is zero then the output voltage is Vin times Xc/(0+Xc)

    You must have an input resistance or the answer will always be Vo = Vin (1)
     
  7. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Isn't your sound card designed to drive an amplified computer speaker that has a 10k ohm input resistance? Then it cannot drive 1.5 ohms which is almost a dead short.

    I think your calculation was very wrong. You also had the resistor and capacitor connected backwards.

    Try it like this:
     
  8. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    Most sound card are able to drive 32ohm headphones directly. At least on the main output. Some sound card may have several output channels. And most often these outputs are designed for around 10K load
     
  9. crutschow

    Expert

    Mar 14, 2008
    12,988
    3,226
    Where did you come up with your formula f=CR? RC is the time constant, which is .33μs for your circuit (how did you get 3030?). The corner frequency (-3dB point) for an RC filter is 1/(2*pi*R*C), thus .22uf and 1.5 ohm have a corner frequency of 482kHz not 3030Hz.
     
  10. c0de3

    Thread Starter Active Member

    May 1, 2009
    50
    1
    Here is what I did... (not that it is right):

    Capacitor: .22uf
    Resistor: 1.5
    .00022 x 1.5 = .00033
    1/.00033 = 3030

    So I was multiplying the capacitance and the resistance and then inverting it.
    This is from the site I was reading:
    So the capacitance value of a given circuit is chosen to allow a certain range of frequencies to pass through. Let's give a few examples: The time constant formula for capacitors is "T=RC" or Time = Resistance x Capacitance. If we take a circuit with a resistance of 10 Ohms, and a capacitance of 10uF, then the Time would be 0.0001 seconds. The Frequency would be 1/.0001, or 10,000 Cycles Per Second ( also called Hertz ). So a 10 Ohm resistor and a 10uF capacitor would yield a circuit passing 10 KHz.
    Here is a direct link to the explanation:
    http://www.electronicstheory.com/COURSES/ELECTRONICS/e101-50.htm

    Looks like that site forgot the *pi part that your formula has!!! Nice. I'll try my filter again tonight, but I'm pretty sure I went the whole way down to 100hz and it was still passing. I'll also follow Audio's suggestion and reverse the cap and resistor.

    Again, thanks for the continued help everyone! I'll post my results tonight.
     
    Last edited: Apr 9, 2012
  11. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The "Electronics Theory" website got the formula completely WRONG and also got the parts positions completely WRONG!

    Why do people post such rubbish?
     
    Sparky49 likes this.
  12. c0de3

    Thread Starter Active Member

    May 1, 2009
    50
    1
    Yeah, it had me completely puzzled why it wouldn't work. And to think he states his site is used as electronics course textbooks! If so, the class must not have a lab section. Too bad this has occurred, as now I don't trust anything I read there and I had liked his writing style.

    Thanks for the help AudioGuru and others! I'm now using: 1/(2*pi*r*c) I'll test that out tonight and see if I have better luck. I'll also make sure my components are in the correct order.


    - C
     
  13. crutschow

    Expert

    Mar 14, 2008
    12,988
    3,226
    .22μf is .00000022F.

    0.00022 is .22mF.
     
  14. Veracohr

    Well-Known Member

    Jan 3, 2011
    550
    75
    0.01uF isn't really "super low", however 1.5 ohms IS.
     
  15. crutschow

    Expert

    Mar 14, 2008
    12,988
    3,226
    Typical capacitors used for audio applications are in the sub nanofarad to several microfarad range, and resistors in the few kilohm to hundreds of kilohm range.
     
    Sparky49 likes this.
  16. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Your math is very wrong again.
    A series 1k resistor then a 0.01uF capacitor to ground produce a lowpass filter with a cutoff at 16kHz.

    For 100Hz, a series resistor needs a 1.6uF capacitor to ground.
     
  17. c0de3

    Thread Starter Active Member

    May 1, 2009
    50
    1
    Actually, I wasn't wrong again... I was wrong still. That quote was taken from before the discovery that I was using the wrong formula! :)

    I was about to report my continued lack of success... however, crutschow has already spotted my error... I was off on my decimal point! Damn the details! :eek:

    I'll try again, tonight and see if I get the expected results.
     
  18. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    A single resistor and single capacitor makes a very simple filter with poor results.
    A 3kHz simple lowpass filter has a flat response up to about 600Hz then its output slowly drops until it has an amplitude of 0.707 times at 3kHz. At 6kHz its amplitude is half of maximum and at 12kHz it is 1/4 of maximum. 24kHz is reduced to 1/8th of maximum. Higher frequencies have their amplitude reduced to half for each octave (frequency doubled). The reduction in level is gradual so that high frequencies are reduced but not eliminated.
     
  19. c0de3

    Thread Starter Active Member

    May 1, 2009
    50
    1
    Thanks for the additional information. I didn't get on the bench last night so I haven't gotten to test my latest plans. I'm sure the information you just gave me would have surprised me on the test bench. I was thinking that at one frequency I'd see a sign wave on the scope and at another it would just be flat... So much to learn. Thanks for the information. I'll still do the test just so I can see it work.

    I'm running into the limits of my very simple wave generation system (PC Sound card) a lot. Do any of you folks have experience with the the cheap generators like this one: HERE (EBay link)? I can't afford one of the fancier models, but this looks like it might work for some simple tests?


    Thank again everyone
     
  20. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The complicated generator you found at E-Bay has fairly high sinewave distortion that might surprise you when it is used to test filters.

    I have seen inexpensive generator kits available.

    I use the free generator program called Audacity.
     
Loading...