Making sense of RC RL LC filters - recognizing high pass versus low pass filters - and a question

Discussion in 'Amp Tech Center' started by peteb, Sep 7, 2019.

  1. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    There were some questions about capacitance filters versus inductance filters, this is what I have found out and what I still want to know.


    A capacitor is a high pass filter, allowing high frequencies to pass and blocking low frequencies, proportionally.


    An inductor is a low pass filter, allowing low frequencies to pass and bocklng high frequencies, in proportion. A choke is an inductor.


    A resistor is the same for all frequencies.




    First order filters are made of a combination of these circuit elements, RC, RL and LC. Each pair can be high pass or low pass, depending on their order. What makes a filter high pass or low pass?



    In each diagram, the input is at the left, and the output is at the dot.






    The most common is the RC filter. The capacitor is high pass. What is note worthy is the low pass filter is made from a high pass capacitor. On the left, high frequencies are passed to ground, leaving low frequencies at the output, making a low pass filter.


    On the right, the capacitor being a high pass filter, allows high frequencies to reach the output.


    9E866AB3-896A-4534-9A8E-A426C2D40194.jpeg



    Next is LR. Similar to RC. The inductor is low pass, so the filter on the left allows low frequencies to reach the out put. The filter on the right is high pass because the inductor passes low frequencies to ground.

    13FB70B3-EB62-4C01-B610-2DCF3B9E7726.jpeg




    Last is LC, common to Fender amps. The LC filter combines the high pass capacitor with the low pass inductor. The filter on the left is low pass because the low pass inductor allows low frequencies to reach the output and the high pass capacitor passes high frequencies to ground, kind of redundant. The filter on the right is high pass because the high pass capacity allows high frequencies to reach the output and the low pass inductor passes low frequencies to ground.



    37CAB832-C20B-4C1B-A7CF-12FADAA2B8A0.jpeg





    Which of these filters are common to fender amps?






    The R, C and L are used to establish the time constant or cutoff frequency for the filter.





    Here is the question





    How do you rate the amount of filtering in an amp? The amount of capacitance? The amount of inductance? The amount of resistance? There are no units for the amount of filtering, are there?
     
    Last edited: Sep 8, 2019
  2. moosie

    moosie Doctor of Teleocity Silver Supporter

    Age:
    62
    Posts:
    15,392
    Joined:
    Jul 18, 2010
    Location:
    Western Connecticut
    I didn't see a question that you didn't already answer... What makes a filter high or low pass depends on the order of the components.
     
    peteb likes this.
  3. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,031
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    First you have to separate the different functions where "filters" are used. Then you sould define the applications where the differing sorts of filter are used, before rating thier parameters.

    The most common "filter" that all tube amps wil have is the ripple filter in the power supply that smooths out the rectified current. This has a common unit for the amount of filtering - percentage of ripple.

    You can also find "filters" tuned to selected frequencies (units of order -1st 2nd 3rd... measured in dB octave) These might be found in feedback loops (look for them in MESA amps NFB)

    Every filter changes phase angle, so you will have another measurement unit - Degrees. .......Plus there are time constants, unit - Seconds

    Between the various stages of the amp you will normaly have a coupling stage with a capacitor, one end is conected to the power supply filters at the plate, and the other end is connected to groung through the grid resistor, forming a CR filter. Much has been written about these stages, and it is no problem to work out the parameters, but who the hell would want to? It has all been worked out decades ago....just grab a .1µ if you want full range with plenty of bass...and use a .001µ if you are going for a high gain bass-cut. In the middle...is in the middle!

    There are plenty of books covering the subject if you want to get into the subject more, or just try WIKI, and if that doesn't sink in to comprehention, then lay-off the subject, and just stick to the conventional component ratings that you can see in our amps.
     
  4. DougM

    DougM Friend of Leo's

    Posts:
    2,967
    Joined:
    Jul 5, 2017
    Location:
    Honolulu, HI
    It's not the order of them, but the way in which they're connected. And, the center frequency at which the components begin filtering is dependent on the impedance of the circuit that they're feeding. This is how crossovers in speakers are designed. A first order crossover is 6db per octave, and has a single capacitor in series with the tweeter and a single inductor in series with the woofer. A second order filter is 12db per octave and has a capacitor in series and an inductor in parallel with the tweeter, and an inductor in series and a capacitor in parallel with the woofer. But, a bandpass filter, instead of just a low pass or high pass, is a lot more complex, like in pedals and amps tone control sections.
     
  5. moosie

    moosie Doctor of Teleocity Silver Supporter

    Age:
    62
    Posts:
    15,392
    Joined:
    Jul 18, 2010
    Location:
    Western Connecticut
    Well, both, no?

    My comment makes less sense now that Pete edited the OP to include the more complex question at the end. I was simply agreeing with his comment:

     
  6. raito

    raito Poster Extraordinaire Silver Supporter

    Posts:
    5,611
    Joined:
    Nov 22, 2010
    Location:
    Madison, WI
    I think you need to look up the term 'transfer function'.

    Of course the RC low pass filter uses the 'high pass' capacitor. It passes higher frequencies to ground instead of the output.
     
  7. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    Thank you for your well informed post. Percentage of ripple is an interesting concept that I am not familiar with.





     
  8. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    Thanks for the responses all. I see people can take this to another level but I am just trying to work out the basics. 1 day ago I could not look at a simple filter and know with reason why it is either a low pass or high pass filter. The most interesting learning is that capacitors and inductors are both filters but they work oppositely of one another, one is high pass and one is low pass.





    Once it is known whether a filter is high pass or low pass, depending on its componant types, the next step is to use the componant values to figure the cut off frequency, and then use that and the fact that the filter is high pass or low pass to evaluate if the filter is effective in its expected function, in this case the question is are 60 cycle and 120 cycle frequencies filtered off as they are meant to be.




    The equations for cut off frequency:


    RC: f = 1 / RC

    LC: f = 1 / 2pi (LC)^1/2

    RL: f = R / L






    Note, see how the resistance is a free agent, working with the C but working against the L.
     
  9. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    One of the underlying questions is what is the affect of having or not having a choke in the power supply? Does it add much filtering compared to RC filters?



    I’m going to look at the effectiveness of RC, LC, and LR filters in the power supply of the narrow panel tweed pro 5E5-A.




    The answer to one question above, which versions of the RC, LC, and LR filters are present in the fender amps?


    It looks to me like the low pass version of each filter type is present in fender amps with chokes.




    BD1B37B5-8737-4B35-A520-452E949C21A2.gif
     
    Last edited: Sep 8, 2019
  10. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    RC filter.



    f = 1/ RC


    R = 10k ohms

    C = 8 uF or 16 uF




    f = 1/10,000 * 16*10^-6 = 6.25 Hz

    f = /10,000* 8*10^-6 = 12.5 Hz



    This is a low pass RC filter, the cut off frequency in the two cases are 6.25 Hz and 12.25 Hz, everything above is filtered off and everything below is passed on, both 60 Hz and 120Hz frequencies would be filtered off.



    Note how both R and C need to be big enough or else the frequency would rise above 60 or 120, making the filter useless.




    How strong is the filter? What does its strength depend on? The value of C? The value of R? The value of C and R?




    I think the strength of the filter is mostly based on the value of the capacitor.


    EDIT:

    And then here is another way to figure it that does include R.



    Doubling C or doubling R will have the same affect on the value of the cut off frequency, doubling either would lower the cut off frequency, lower and further away from 60 Hz. I think if the cut off frequency is lowered further away from 60 Hz, the filter will be more effective reducing 60 cycle noise.
     
    Last edited: Sep 8, 2019
  11. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    LC filter



    f = 1 / 2pi (LC)^1/2


    C = 16 uF

    L = 4 Henry’s (Fender?) or 40 Henry’s (Marshall?)



    f = 1 / 2pi (4*16*10^-6)^1/2 = 20 Hz


    f = 1 / 2pi (40*16*10^-6)^1/2 = 6.3 Hz





    This is a low pass LC filter, the cut off frequency in the two cases are 20Hz and 6.3Hz, everything above is filtered off and everything below is passed on, both 60 Hz and 120Hz frequencies would be filtered off.



    1. Note how both L and C need to be big enough or else the frequency would rise above 60 or 120, making the filter useless.




      How strong is the filter? What does its strength depend on? The value of C? The value of L? The value of C and L?




      I think the strength of the filter is mostly based on the value of the capacitor. The inductor is in the role of the resistor in the RC filter and not really acting as a filter in its own right. By this I mean the inductor is low pass filter and both 60 Hz and 120 Hz are low frequencies. The inductance itself is likely to pass 60 Hz.




      And then here is another way to figure it that does include L.



      Doubling L or doubling C will have the same affect on the value of the cut off frequency, doubling either would lower the cut off frequency, lower and further away from 60 Hz. I think if the cut off frequency is lowered further away from 60 Hz, the filter will be more effective reducing 60 cycle noise.



     
  12. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    Then there is LR filter, this is the odd one out.


    f = R / L



    Note: see how the R has moved to a place in the equation not ever occupied by L or C, and R in the RC equation was in the denominator.




    R = 10,000 ohms

    L = 4 Henry’s (Fender?) or 40 Henry’s (Marshall?)




    f = 10,000 / 4 = 2,500 Hz


    f = 10,000 / 40 = 250 Hz



    This is a low pass LR filter which will pass all frequencies below 2,500 Hz or 250 Hz including both 60 Hz and 120 Hz. Out of the three first order filters of LRC types, this is the only one where increasing a componant value, in this case the value of R, will raise the cut off frequency. Raising the cut off frequency of a low pass filter makes the low pass filter weaker. This is the one case out of these three that raising a componant value, the resistor, lowers the strength of the LR low pass filter.




    This shows that adding resistance can add or subtract from the strength of the filtering, and in the case of the low pass LR filter, the large size of the 10K resistor makes the filter ineffective.
     
  13. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    Findings



    Capacitors which are high pass filters themselves, provide the majority of the filtering even in amps with inductive chokes. Every node has a cap to ground passing high frequencies to ground, this is the general synopsis of how all power system filtering is done in Fender amps.



    Resistors are not filters themselves but in the case of the common RC low pass filter, the resistor leverages the filtering capability of the capacitor making for a more effective or stronger filter. Increased Resistance can also lower the effectiveness of the RL low pass filter.



    the mystery is how the choke provides increased filtering, which it must, or why would Fender go to this great expense.

    If there is a choke in the amp, RL filtering is present but it has no effectiveness. If there is a choke, it is used with a cap in a low pass LC filter. The choke is not the filter, the cap is. The choke acts in an LC filter as a resistor acts in an RC filter, the impedance of the inductance leveraging the capacitor to make a more effective filter.





    I am fairly certain that these are the fundamentals. If they are wrong please let me know.




    After all of this I still don’t understand the importance of the choke. Anyone??





    Back to the original question. How to rate the strength of a filter?




    My Best guess:


    1. the value of the capacitance at node 1
    2. The sum of the capacitance at all nodes
    3. The amount of inductance in the choke.





    The unit for filtering might as well be Farads.



    Edit:



    The measure of the strength of a filter could include the resistance of RC and the impedance of LC.

    The units could the resistance or impedance multlied by the capacitance, the units being Farad Ohms.


    But wait, the initial filter cap, or primary node would have zero Farad ohms, because it has no resistor before it but the internal resistance of the rectifier and cap might apply.
     
    Last edited: Sep 8, 2019
  14. gusfinley

    gusfinley Tele-Holic

    Posts:
    843
    Joined:
    Jun 11, 2014
    Location:
    SLC, UT
    So far you have looked into series resistive-divider-like circuits where one of the "resistors" is replaced with a reactive component ( capacitor or inductor )

    Don't forget about parallel filters where a reactive component is strapped in parallel a resistor, though these are a little more difficult to find information about.
     
  15. gusfinley

    gusfinley Tele-Holic

    Posts:
    843
    Joined:
    Jun 11, 2014
    Location:
    SLC, UT
    The "strength" of a filter is rated in terms of its:

    a) Frequency Response

    b) Phase Response

    These two parameters are shown visually by a Bode Plot:

    [​IMG]

    The Bode Plot - or Filter Response - can be applied for each filter network or as a combination of filter networks.

    The mathematical relationship between the output of the filter and the input of the filter is know as the "Transfer Function" mentioned previously in this thread.

    Many cheap digital modeling circuits are created by simply shoving a known signal into a filter network ( amplifier ), measuring the frequency response the the output and that use that information to create a Digital "Transfer Function" that gets "close enough." This is limited by the processing power of the DSP and by the amount of memory available to the DSP. Then EQ is applied post-transfer-function to tweak the output to taste.

    More complicated digital models will model the effect of each component and apply a modified transfer function as the settings are changed. This obviously costs more in DSP power and memory and is therefore more expensive to produce.

    There are also many varieties of modelling in between these extremes.
     
    W.L.Weller and peteb like this.
  16. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia

    Good point






    And this too.



    M main goal was to be able to look at a schematic, ID the critical parts that affect filtering and make an estimate of the strength of the overall filtering.



    This thread was in response to another thread where the benefit of a choke was questioned. I feel like I have a good idea how these filters work, but there is still the question about how beneficial a choke is, and more importantly how a choke improves filtering.
     
  17. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,031
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    You title the thread "Making sense of RC RL LC filters - recognizing high pass versus low pass filters - and a question"

    QUOTE:The most interesting learning is that capacitors and inductors are both filters but they work oppositely of one another, one is high pass and one is low pass. END QUOTE

    No! Both can be used in creating filters, where one blocks AC the other DC....not high and low.

    QUOTE: Note, see how the resistance is a free agent, working with the C but working against the L. END QUOTE

    Nonsence.


    I'm sure you probably drew many other false conclusions in your ramblings, but I can't be bothered reading so much tosh.


    But just to pick-up on the last comment that I see above;

    QUOTE: I feel like I have a good idea how these filters work, but there is still the question about how beneficial a choke is, and more importantly how a choke improves filtering. END QUOTE

    If have a feeling you have a good idea how these filters work, why do you ask about the choke?

    CLUE: The use/function of the choke in an amp is not just as a component to improve filtering.....now go read a book about the subject.


    (WHY ISN'T THE EDITING AND POSTING FUNCTION WORKING PROPERLY?)
     
    Last edited: Sep 10, 2019
  18. kbold

    kbold Tele-Holic Silver Supporter

    Posts:
    621
    Joined:
    Jul 21, 2015
    Location:
    Australia
    A cap is frequency dependent; blocks low frequencies/passes high frequencies
    An inductor is also frequency dependent; blocks high frequencies/ passes low frequencies.
    So both are filter elements.
    Use both and you increase the effectiveness of the filter, as well as allowing selective filtering (eg band-pass, band-stop)
     
  19. peteb

    peteb Friend of Leo's

    Posts:
    4,081
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    thank you Kbold,


    what you wrote is on the same subject as this last part I wanted to add as my conclusion.




    The choke is a filter but it appears that the inductance of the choke is not used as a filter.



    Rather, the impedance of the inductance of the choke is used much as the resistance of the resistor to assist and modify the filtering behavior of the capacitor.




    I now agree with the others in the other thread. The function of the choke, even though the choke is a filter, is more like the function of a resistor in an RC choke, making a resistor the direct replacement for a choke instead of a filter cap being the direct replacement of the choke.
     
  20. Bendyha

    Bendyha Friend of Leo's Silver Supporter

    Posts:
    2,031
    Joined:
    Mar 26, 2014
    Location:
    Northern Germany
    .
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.