RF Low Pass Filter

Discussion in 'General Electronics Chat' started by Mazaag, Sep 24, 2007.

  1. Mazaag

    Thread Starter Senior Member

    Oct 23, 2004
    255
    0
    Hi guys,

    Just wanted to ask if an RC Low Pass filter (100Hz cut-off) is sufficient to filter out high frequency components of a signal coming in at 460KHz without having any inductive effects come into play ? Or are there special "RF" filters that should be used in this case ?

    Thanks
     
  2. nanovate

    Distinguished Member

    May 7, 2007
    665
    1
    Yes it should work.
     
  3. Mazaag

    Thread Starter Senior Member

    Oct 23, 2004
    255
    0
    would using a non-inductive resistor make any difference? would it be neccessary?
     
  4. nanovate

    Distinguished Member

    May 7, 2007
    665
    1
    What is the application? You could use a ferrite bead in series in addition for the >460KHz. What type of capacitors do you intend to use. Use a few in parallel -- tantalum and ceramics (X7R min, NPO preferred).
     
  5. Mazaag

    Thread Starter Senior Member

    Oct 23, 2004
    255
    0
    Thats the question, i'm not entirely sure what to use ..

    I have a signal which is a sinusoid running at 460KHz and I need to check for low frequency components..(below 100Hz) so Ineed to filter at that frequency...
     
  6. nanovate

    Distinguished Member

    May 7, 2007
    665
    1
    So you only care about the stuff below 100Hz? Or do you need to keep the 460KHz signal also?
     
  7. Mazaag

    Thread Starter Senior Member

    Oct 23, 2004
    255
    0
    I jsut care about the stuff below 100Hz... this filter is basically an Anti-Aliasing filter preceeding a Digital Oscilloscope..
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Take a look here:
    http://www.arrl.org/tis/info/pdf/9902044.pdf

    Low-pass passive RF filter, 50 Ohm input/output, cutoff freq around 18 MHz, down 60 dBm by 28 MHz. You could probably improve on that by adding shielding between the stages, but that will change your coupling between the stages.

    You'll need to tune it, preferably using a network analyzer with an S-parameter test set. You COULD use a spectrum analyzer and a frequency sweeping signal generator; for that matter, it's been done with a manual-cranked sig gen and a VOM. You have to be somewhat of a masochist though ;)

    Yes, the 18 MHz band edge is quite a bit higher freq than you were asking for - but if you really want to filter that low, you're going to need some REALLY large inductors. There are values supplied in Table 1 to get the roll-off to start @ 1 MHz.

    Use high-Q caps. Watch the Q on your inductors - if you try to over-squeeze the coils, you'll kill their Q and wind up with lousy insertion loss. :( It can be tough to get good toroids - they can be pretty inconsistent. Baking them for 24 hours @ 100°C just prior to tuning the filter will ensure they're pretty dry. Yes, moisture will get into the toroids if you scratch them, and yes, it will throw your filter out of tune.

    You COULD build an active Sallen-Key filter using an op amp and some components - but then you'd need power supplies, have to deal with S/N ratios and all that other mess. A well-designed and properly tuned passive LC filter should do the trick for you quite nicely.
     
  9. Mazaag

    Thread Starter Senior Member

    Oct 23, 2004
    255
    0
    Thanks for you reply guys
     
  10. nanovate

    Distinguished Member

    May 7, 2007
    665
    1
    Down at 100Hz you are probably better off with an RC filter augmented with a balun or ferrite bead. I am assuming that the 3dB point is at 100Hz. What is your stopband desired attenuation? What does the 100Hz signal look like?
     
Loading...