bandwidth filter

Discussion in 'General Electronics Chat' started by fred_uk, Feb 8, 2009.

  1. fred_uk

    Thread Starter Active Member

    Jan 3, 2009
    67
    0
    can anyone help with the design of a bandwidth filter, i have the documentation but cant seem to work out the values, I am a newbie at this.

    the freqency is 500hz to 3000hz

    Thanks
     
  2. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Have you chosen a filter circuit topology?

    hgmjr
     
  3. mik3

    Senior Member

    Feb 4, 2008
    4,846
    63
    Are you interested in active filters?

    Where is the documentation?
     
  4. fred_uk

    Thread Starter Active Member

    Jan 3, 2009
    67
    0
    sorry guys, i must admit i know very little about filters, other than i need a bandpass filter, from 500hz to 3000hz, i was then hopeing to use a transitor to provide a switched output on the presence of a tone within the bandwidth specified

    sorry about the lack of documentation, im a newbie at this
     
  5. mik3

    Senior Member

    Feb 4, 2008
    4,846
    63
    Then you can use a band pass filter like this one here:

    http://www.captain.at/electronics/active-filter/

    Then use a comparator to compare the output of the filter with a DC voltage and when the output of the filter is greater than the DC voltage the comparator will go high.
     
  6. fred_uk

    Thread Starter Active Member

    Jan 3, 2009
    67
    0
    thanks mike,

    will i be able to cover the full bandwidth (500hz to 3000hz) in which case the centre freq would be 1750hz ?? would that be correct?
     
  7. mik3

    Senior Member

    Feb 4, 2008
    4,846
    63
    You if you set the bandwidth of the filter properly. However, remember that filters are not ideal they will pass frequencies near the bandwidth with a respectable amplitude.
     
  8. Skeebopstop

    Active Member

    Jan 9, 2009
    358
    3
    Hey Mike, do you know of a site showing the derivation of the filter you posted?

    I found this one which seems easier and simpler to implement (See inverting band pass):

    http://www.electronics-tutorials.ws/filter/filter_7.html

    Perhaps it doesn't have as good an attenuation?

    One thing about your link Mike, I popped in some values and everything looked ok except R3, it calculated me a - value for resistance!? In other pages I found implementing the same filter, they just didn't have R3. Any reason it is there?
     
  9. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The ratio of (R4/R3) +1= gain of the opamp. Remove R3 then the opamp has no gain.

    This extremely simple "bandpass" filters have almost no attenuation of outside signal frequencies. They attenuate frequencies you want because its cutoff frequencies are droopy and are not sharp.
    The better filters I showed on the other website are "2nd-order" and have twice as much attenuation of outside signal frequencies as these 1st-order filters.
    3rd-order and 4th-order filters are even better.

    You asked why you can still hear 100Hz when the highpass filter is 4th-order. Of course, the attenuation is not much, you need 10th-order.
     
  10. leftyretro

    Active Member

    Nov 25, 2008
    394
    2
    My gut says you would have better luck, more realistic component values and better circuit Q if you designed it in two stages, a lowpass active filter stage driving a highpass active filter. Most bandpass filter don't have to deal with such a wide bandpass as related to it's center frequency.

    Lefty
     
  11. fred_uk

    Thread Starter Active Member

    Jan 3, 2009
    67
    0
    hi guys,

    sorry another bone question to ask...

    what chip am i using to create this filter??
     
  12. Skeebopstop

    Active Member

    Jan 9, 2009
    358
    3
    I was referring to Mike's circuit. -R2/R1 is the gain of the inverting amp. So what the is R3 doing? I found a site which had the same bandpass inverting filter but didn't have that R3 there.
     
  13. Skeebopstop

    Active Member

    Jan 9, 2009
    358
    3
    Can you even get 10th order in hardware? I thought that was purely a DSP thing.
     
  14. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
  15. fred_uk

    Thread Starter Active Member

    Jan 3, 2009
    67
    0
    what do you guys think of this circuit, a high pass filter followed by a low pass and then a comprator circuit. All i need to do is light the led when a tone between 500hz and 3khz is detected, however my maths isnt veryy good and i could do with some help on the values of components.

    what do you think of the circuit design so far. I have been told it will need to run off two 9 volt batteries to provide a simple bi-polar power supply.
     
  16. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    You will need to rectify afther the filters to have a DC voltage that you can compare with the DC reference.
    What type of filters do you want to use ?
    Depending on the resistor and capacitor values the filter will have different properties.

    Greetings,
    Bertus
     
  17. fred_uk

    Thread Starter Active Member

    Jan 3, 2009
    67
    0
    what do you mean by what type of filters? i thought they were high pass and low pass, am i missing a bit to my comparator circuit?
     
  18. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Mike's circuit is a simple narrow-bandwidth bandpass filter.
    Your requirement has a bandwidth that is too wide for it.
    At frequencies that you want high attenuation it barely has any attenuation.

    Look it up in Google: Multiple Feedback bandpass filter Circuit
     
  19. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Your comparator compares the DC output voltage of the filter IC to the setting of the trimpot:
    Turn the pot up and the LED lights.
    Turn the pot down and the LED turns off.

    You need to rectify the output of the filter, filter the pulses with a capacitor then drive a transistor with the result. The transistor will turn on the LED when there is a signal.

    Opamps do not need a dual-polarity supply if their input is biased at half the supply voltage.
     
  20. fred_uk

    Thread Starter Active Member

    Jan 3, 2009
    67
    0
    so would the one circuit you listed above be suitable to such a wide bandwidth??

    or would i need more than one centred on different frequencies

    the tones are at:

    500hz

    750hz

    1000hz

    2000hz

    3000hz

    thanks
    fred
     
    Last edited: Feb 10, 2009
Loading...