9th order low pass butterworth filter

Discussion in 'General Electronics Chat' started by ajaysathya2604, May 29, 2018.

  1. ajaysathya2604

    Thread Starter New Member

    May 29, 2018
    9
    0
    Hello,

    I am designing a 9th order low pass butterworth filter. I have created the design at the moment. I would like to Know if an addition of a 5th and 4th order IC low pass filter would create a 9th order filter or should it be 2+2+2+2+1 like the usual way. I would like to use ICs available in the market to make this filter.
     
  2. crutschow

    Expert

    Mar 14, 2008
    23,112
    6,853
    Combining two or more Butterworth filters does not give you a Butterworth filter.
    All the filters in the chain have to be configured to give the proper Butterworth response at the output.

    You could likely use one of these.
     
  3. ajaysathya2604

    Thread Starter New Member

    May 29, 2018
    9
    0
    Thank you for the link.

    Yes, I understand that. Im looking forward to make the design as compact as possible.
     
  4. danadak

    Distinguished Member

    Mar 10, 2018
    3,497
    785
  5. ebeowulf17

    AAC Fanatic!

    Aug 12, 2014
    2,934
    641
  6. OBW0549

    Distinguished Member

    Mar 2, 2015
    2,948
    2,706
    Keep in mind that while synthesis of a 9th-order filter, such as what you show in the diagram attached to your post, is rather easy (given appropriate filter design software), actual implementation of the design will be extremely difficult if not outright impossible due to the requirement for either extremely tight-tolerance components (which are expensive) or a lot of labor-intensive trimming to get the desired response. Or both.
     
  7. danadak

    Distinguished Member

    Mar 10, 2018
    3,497
    785
    For sure the solution I posted earlier is only good to < 100Khz. Ignore
    that post. The DSP solution for a Mhz type of filter would be the advanced
    TI DSP parts, or Analog Devices or FPGA solutions.

    Regards, Dana.
     
  8. SLK001

    Senior Member

    Nov 29, 2011
    1,440
    591
    Building a 9th order filter of any kind is near impossible - especially for a beginner.
     
  9. Papabravo

    Expert

    Feb 24, 2006
    12,284
    2,723
    Having the test equipment to verify the filter is near impossible for a beginner as well. A multi-meter just won't do the job.
     
    ajaysathya2604 likes this.
  10. DECELL

    Member

    Apr 23, 2018
    96
    19
    One less amp to worry about:
    upload_2018-5-30_1-26-48.png
    Sim this! upload_2018-5-30_1-43-13.png
    digital pah!
     
    ajaysathya2604 likes this.
  11. danadak

    Distinguished Member

    Mar 10, 2018
    3,497
    785
  12. ajaysathya2604

    Thread Starter New Member

    May 29, 2018
    9
    0
    So, from the comments I understand that my simulated filter design is really hard to implement and receive results similar to that of the simulations. Would it be a optimal to add a passive filter as mentioned above? If yes, what would be the setbacks?
     
  13. ajaysathya2604

    Thread Starter New Member

    May 29, 2018
    9
    0
    Totally understandable
     
  14. DECELL

    Member

    Apr 23, 2018
    96
    19
    The active filter need some very fast opamps and close attention to layout an it is unlikely to perform well on a breadboard. Stray capacitance and inductance will make your life hell!
    If you go with the passive you can easily set the input and output impedance to suit your application.
    You may want to relax on pass band ripple to keep it simple.
    Sticking to preferred component values gives you a good enough result most of the time. eg:
    upload_2018-5-30_15-20-11.png
     
    ajaysathya2604 likes this.
  15. danadak

    Distinguished Member

    Mar 10, 2018
    3,497
    785
    Pros and cons active vs passive filter design -

    1) - No Ls required (generally), active.
    2) - Generally smaller design volume wise, lower weight, mass, active.
    3) - Lower Noise (you have to do the analysis, this result is not chiseled in stone), passive.
    4) - Less components (generally), passive.
    5) - No external power needed, passive, although you may need an amp to compensate for losses.
    6) - No nonlinearities,. passive, but microphonics in magnetics can present as noise, non linearity.
    7) - Higher and lower temp operation, passive.

    I am sure there are more.

    Regards, Dana.
     
    Last edited: May 31, 2018
    ajaysathya2604 likes this.
  16. ajaysathya2604

    Thread Starter New Member

    May 29, 2018
    9
    0
    Based on all the inputs from you amazing people, I think implementing a passive filter would be highly feasible at the moment.

    Thank you :)
     
Loading...