1. Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.
    Dismiss Notice

Guitar Audio Filter For Frequency Detection

Discussion in 'Electronic Projects Design/Ideas/Reviews' started by transistance, Jul 8, 2008.

  1. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    [​IMG]
    here is the filter that I'm having problem with. This thread is as a follow up from "2 x TL071 = TL072??". I can not get any audio after the comparator, am i not supposed to? i remember audioguru tellin me i should get a very distorted sound, yet that was with schmitt trigger.

    also the original design had a 16u cap for C5, is replacing it with a 10u going to make a huge difference? (btw just to make it clear C4 and C5 are electrolytic)

    thanks in advance,

    P.S. sorry for the crappy schematic, i had to use pspice student version at work...
     

    Attached Files:

    Last edited: Jul 8, 2008
  2. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    what's the reason behind supplying 2.5Vdc to positive inputs of the TL071's btw? This circuit was redesigned with reference to nilS thegimmick btw.. and he referenced it off some german guy...
     
  3. ericgibbs

    ericgibbs Well-Known Member Most Helpful Member

    Joined:
    Jan 4, 2007
    Messages:
    21,235
    Likes:
    645
    Location:
    Ex Yorks' Hants UK
    ONLINE
    Replied to your PM.:)
     
  4. dave

    Dave New Member

    Joined:
    Jan 12, 1997
    Messages:
    -
    Likes:
    0


     
  5. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul

    i'll paraphrase ericgibbs' pm here so ppl like me have an idea:

    ericgibbs:
    "R2/R3 divide the +V1 into half [the centre point of the +V supply] By applying this Vss/2 to the TL opa NI inputs , this sets the operating point of the opa's OUTPUTS to V1/2. This means without an audio signal the opa outputs sit at +2.5V, so any audio signal will drive the opa down towards 0V and up towards +V1.. This means the signal will be symetrical about +2.5V,,, this is what you want in a 'ac' amplifier, when its only got a single power supply line [+5v].

    The C4 is just noise decoupling of this centre point voltage.

    C5 will act as a filter with with R9/10. The LM393 comparator has positive feedback R12, so R9/10 C5 and R12 work together to determine at which voltage point on the audio signal, the signal is squared up by the LM393.
    "
    This clears a lot of stuff up for me; except one :eek:. If anyone could explain how the comparator and that part of the circuit works? with a little equations incorporated if possible.

    your input is greatly appreciated,

    (i feel like an information leech, i haven't been able to really help anyone yet. but it'll happen as soon as i get this project done, I'm going to post a huge tutorial on automated guitar tuning with servos)
     
    Last edited: Jul 8, 2008
  6. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,573
    Likes:
    950
    Location:
    Canada, of course!
    ONLINE
    Why are there so many threads about this circuit. I wish it is in a single thread so I can see the original schematic. I searched and found the original schematic.
    1) the value of C3 is 453 times too high.
    2) The minimum supply voltage for a TL071 is 7V. This circuit has only 5V. The opamps might not work.
    3) A guitar pickup is designed to be loaded with 1M to 3M. Usually a vacuum tube or FET is used. This circuit has 4.7k which is like a dead short and will reduce the output from the pickup a lot.

    If the capacitor value is corrected, if the opamps work and if the pickup has enough output then of course the output of the comparator is a square-wave. The comparator is a schmitt-trigger.

    The value of C5 is fine if it is 10uF.
     

    Attached Files:

  7. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    lol.. kk that was a typo, but i didnt know about the 7v supply, I'm going to try that soon.. thx audioguru
     
  8. ericgibbs

    ericgibbs Well-Known Member Most Helpful Member

    Joined:
    Jan 4, 2007
    Messages:
    21,235
    Likes:
    645
    Location:
    Ex Yorks' Hants UK
    ONLINE

    By posting my reply to your PM, without posting your PM question to me, has made my reply look nonsense, as its out of context..

    If you must post members replies to your PM's, please also post the question you asked in the PM.
     
  9. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,573
    Likes:
    950
    Location:
    Canada, of course!
    ONLINE
    Eric explained how the comparator works:
    A comparator switches its output low when the voltage on its (+) input is lower than the voltage on its (-) input.
    The output goes high when the voltages are reversed.

    R12 provides positive feedback together with R10 for a snap-action Schmitt-trigger.
     
    • Like Like x 1
  10. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    I rebuilt the circuit with +7V power supplies to LM393 and TL071's which got my circuit working!! But I'm getting some inconsistencies; such as a 100-120Hz and sometimes 600-620Hz square wave when there is no input. The range that I'm interested in is 50-380Hz roughly.

    First of all, I'm thinking that this circuit is a 3rd order low pass filter, so how is a 600+Hz signal is even passing through to output?

    Secondly,when I get the 100-120Hz noise it's a big problem because it's in the range of frequencies that I'm trying to analyze. When I generate ~82Hz (low E) recorded guitar signal, the oscilloscope sees about 160Hz.

    FYI - I do not own a proper oscilloscope, I use my laptop and Scope software, with coaxial cables coming out from headphones as signal generator and mic as scope. The coaxial cable had solved my noise problems previously coming from my magnetically crowded room.

    What can I be doing wrong? My friend suggested that I move my circuit to a PCB since the protoboard's grid might collect a lot of EM, but I'm still in the prototyping phase!

    thanks in advance,

    -I would like to thank audioguru and ericgibbs for their help to get me up to this point.
     
    Last edited: Jul 10, 2008
  11. ericgibbs

    ericgibbs Well-Known Member Most Helpful Member

    Joined:
    Jan 4, 2007
    Messages:
    21,235
    Likes:
    645
    Location:
    Ex Yorks' Hants UK
    ONLINE
    hi,
    Got your PM.:)
    What power rail decoupling do have on your project board.?

    Is it possible to post a picture of the project board, so that we can get some idea on how its laid out.

    Whats the mains frequency in Istanbul.?

    EDIT: did you get the 10uF/16uF
     
    Last edited: Jul 10, 2008
  12. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul

    I'm not too familiar with what power rail decoupling is supposed to be like exactly. I have one rail fed directly from a power supply ~7-8V, other rail going through a 7805, grounds are connected to each other. Is power rail decoupling like a capacitor between ground and power to get rid of AC? I think I just placed a 10uF electrolytic for the 5V supply, and nothing for the 7V.. oops i guess?? What values should I use?

    I will post a picture when I get home. I will this time... like for real! :)

    Istanbul's electricity is distributed at 220VAC @ 50Hz.

    I went with the 10uF electrolytic instead of 16uF.
     
    Last edited: Jul 10, 2008
  13. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,573
    Likes:
    950
    Location:
    Canada, of course!
    ONLINE
    The cable from the gutar to the circuit must be shielded audio cable to avoid mains hum pickup.

    I have never seen a second-order lowpass filter like yours. It might be a simple bandpass filter if it was wired a little differently.
    Here are two second-order lowpass filters:
     

    Attached Files:

  14. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    Just to clarify things; here is what I have, and what I'm experiencing with my circuit:

    Updated schematic
    Image of circuit
    Static noise my computer stopped analyzing to the right of the screen here because it slowed down due to print screen command
    Noisy output 81.81 Hz btw, sorry for the crop
     

    Attached Files:

    Last edited: Jul 10, 2008
  15. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,573
    Likes:
    950
    Location:
    Canada, of course!
    ONLINE
    There are a few guitar runer circuits on the internet. One has a real Sallen and Key Butterworth lowpass filter and a 1M input impedance.

    I think your circuit has an input impedance that is far too low and I don't think the filter is correct.
     
  16. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    [​IMG]

    something like this ^^ u say? This one is for 5 string bass it says but, I'm gonna check out the frequencies attenuated maybe, it can be used for a 6 string electric guitar...
     
    Last edited: Jul 10, 2008
  17. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    I also came across another design, the schematic was hard to read so I redrew it, the values might be incorrect. If they seem out of place please let me know. Is this the kind of filter + comparator combo, I should be going for??
     

    Attached Files:

  18. Nigel Goodwin

    Nigel Goodwin Super Moderator Most Helpful Member

    Joined:
    Nov 17, 2003
    Messages:
    39,324
    Likes:
    653
    Location:
    Derbyshire, UK
    That certainly looks a lot more plausible, with an input suitable for a guitar.
     
  19. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    do all values in the pspice schematic above look easily obtainable?

    As in are the values of resistors and capacitors that are available in electronics stores?

    For the 2.5v power supply-I think feedback?- connecting to U4 opamp, can i just use any value R's for a voltage divider on 5V to achieve 2.5V, or would the resistors screw up the response of the last opamp - which i think is functioning as a comparator? - ?

    I think I should also move to a quad opamp to save some space since this is supposed to fit inside a guitar.
     
    Last edited: Jul 11, 2008
  20. audioguru

    audioguru Well-Known Member Most Helpful Member

    Joined:
    Mar 16, 2004
    Messages:
    32,573
    Likes:
    950
    Location:
    Canada, of course!
    ONLINE
    Tour schematic was saved as a fuzzy JPG file type instead of as a very clear GIF or PNG file type.
    Your batteries are drawn with backwards polarity.

    You changed everything:
    1) The original circuit has a 1M input impedance. Yours is 500k.
    2) The original circuit uses a 1uF capacitor in the gain setting parts for the 1st opamp for a low cutoff frequency of 16Hz. Your 50nF capacitor for C2 has a low cutoff frequency of 320Hz. It will have a flat response down to 1.6khz.
    3) The original circuit has C4 double the value of C3 to make a sharp Butterworth response on its lowpass filter. Yours has equal values which makes a droopy Bessel filter.
    4) The original circuit uses 100nF for C5. Yours has only 10nF.
     
  21. transistance

    transistance New Member

    Joined:
    Apr 3, 2008
    Messages:
    154
    Likes:
    1
    Location:
    Istanbul
    Self Intervention

    I did some calculations on my own this time to create my filters. The real important one is the low pass. I think the capacitors I have chosen will make a sharper curve. The high pass filter is there as a precaution and serves as a barrier to only pass through the frequencies I am interested in. The rest of the circuit has been taken from 5-string bass tuner from electricdesign that u can find at the bottom of the first page of this thread.

    I hope that this will be an effective design, I took care to have high input impedance and to choose the right sized capacitors for my low pass filter to get a sharper curve.

    What I learned from this project is that one is better off understanding the filter concept and creating his own; makes everything so much clearer. Also, boxing and labeling separate parts of your design helps you understand the flow mechanism.

    Now, I am TRULY ready for your already appreciated inputs.

    thanks in advance,
     

    Attached Files:

    Last edited: Jul 14, 2008

Share This Page