Is it possible to create a 0.001Hz bandwidth low pass filter?

Discussion in 'General Electronics Chat' started by richiechen, Dec 14, 2012.

  1. richiechen

    Thread Starter Member

    Jan 1, 2012
    93
    0
    Hi ALL

    I am trying to build a phase lock loop. And at the end of the PLL will be a low-pass filter, which only allows DC to pass. The DC voltage will be measured by an ADC.

    Because the narrower the filter, the less the noise will be measured by ADC.

    So here is my question, is it possible to create a 0.001Hz low pass filter, which only allows DC to pass??

    Thank you very much.

    Regards
    Richie
     
  2. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    It will take a very long time (13.3 minutes) to charge or discharge.
     
  3. bertus

    Administrator

    Apr 5, 2008
    15,638
    2,343
    Hello,

    The lower you make the loop filter frequency, the longer it takes to react on changes in frequencies.
    The reaction time is 5 * Tau.
    A 0.001 Hz has a delay of 5 * 1/0.001 = 5000 seconds.

    Bertus
     
  4. Papabravo

    Expert

    Feb 24, 2006
    10,135
    1,786
    It is theoretically possible, but probably impractical. For example a passive 8-pole Butterworth for that frequency would require inductors and capacitors with impractical values.

    You might be able to get there with a Sallen-Key opamp based one, but as has already been pointed out there are tradeoffs.
     
  5. MrChips

    Moderator

    Oct 2, 2009
    12,414
    3,353
    Imagine an RC filter with 1MΩ series resistor and 1000μF capacitor to ground!

    This would take 1000s to reach 67% of the input voltage, about 1 hour to settle to within 2% of the input voltage.
    Your PLL would be useless.
     
    Last edited: Dec 14, 2012
  6. bertus

    Administrator

    Apr 5, 2008
    15,638
    2,343
    richiechen and ifixit like this.
  7. #12

    Expert

    Nov 30, 2010
    16,247
    6,744
    It seems you have talked yourself into focusing on a method that is not practical. Go with the recommendation from Bertus and educate yourself until you realize a different method will work.
     
  8. Jaguarjoe

    Active Member

    Apr 7, 2010
    770
    90
    About 25 years ago I worked with a guy named Vern Siegel. He made a filter with a two week time constant. It had to do with earth tremors and the like. Vern is still out there, if you google his name it should be trivial to get in contact with him and go from there. Beware- he is incredibly smart.
     
  9. ifixit

    Distinguished Member

    Nov 20, 2008
    638
    108
    Hi richiechen,

    Digitize your signals and implement as many of the PLL functions in software as you can. It is not practical to implement a 1 millihertz low pass filter in hardware, but very easy in software.

    What is the purpose of this PLL anyway?

    Good Luck,
    Ifixit
     
    richiechen likes this.
  10. richiechen

    Thread Starter Member

    Jan 1, 2012
    93
    0
    Thanks Ifixit!!
    Your advice is quite practical. But in current version I am trying to build a PLL in hardware.

    The purpose of the PLL is to measure a sine wave of tens of nano volt amplitude.

    Best regards
    Richie CHEN
     
  11. richiechen

    Thread Starter Member

    Jan 1, 2012
    93
    0
    Thank you all for the useful information.

    The sine wave I am trying to lock in is 1kHz actually. The reason for me to use a very low band LP filter is because I want to suppress the nosies. I am trying to detect a sine wave of tens of nano volt amplitude. So I guess the noise is a quite critical problem.

    Any idea of suppressing the noises?

    Thank you all very much.
     
Loading...