Clicking in audio

Discussion in 'General Electronics Chat' started by rbkxiiowe, Apr 15, 2009.

  1. rbkxiiowe

    Thread Starter Member

    Mar 13, 2009
    15
    0
    Hi. I'm trying to make a simple tremolo effect for a guitar by using a square wave oscillator to 'wobble' the gain of an amplifier that the guitar's signal passes through.

    I'm using a 555 timer to produce the square wave which then goes into a homemade optoisolator (LED, LDR, and black tape). This is then connected to pins 1 and 8 of an LM386 to wobble the gain.

    The guitar goes into the LM386 and then out into a separate guitar amp.

    This does give the tremolo effect I wanted but my problem is that there is a clicking sound from my guitar amp in time with the oscillations. What's causing the clicks and how do I get rid of them?

    I've attached a diagram of my circuit.

    Cheers.
     
  2. PRS

    Well-Known Member

    Aug 24, 2008
    989
    35
    I've never made a tremolo, but I see the reasoning. I think the clicks could be due to the finite rise and fall time of the square wave. Try a sine wave.??? Use a simple RC oscillator, varying the frequency with the resistor. This resistor would give you the amount of tremulo you desire.
     
  3. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    That or use a RC filter after the 555. Select the rolloff, use 2 in a row, and you will get a rough approximation of a sine wave.
     
  4. rbkxiiowe

    Thread Starter Member

    Mar 13, 2009
    15
    0
    Thanks you two. I'm very new to this electronics stuff but I'll have a go myself but I might come back here with questions like "RC filter - how???"

    I'll give it a go.
    Cheers.
     
  5. Darren Holdstock

    Active Member

    Feb 10, 2009
    262
    11
    Sounds like classic feedthrough from the edges of the square wave - the sine wave PRS suggests shouldn't give such problems. The LM386 isn't designed to have the gain inputs used for modulation (you'll want a transconductance amp like the LM13700 for stuff like that) so it's not behaving abnormally under the circumstances. The datasheet also has a 10 uF cap in series with the external gain-setting resistor - you might want to try slipping one of those in to see if that helps.

    Minor point - tie down the In and Out lines with a 1 M resistor to ground, because if they're also unterminated at the other end then the signal line on the lead will carry some DC voltage from capacitor leakage, and then the cable will crackle like hell when moved or trodden on.

    That said, it's an ingenious circuit; keep up the fine work rbkxiiowe.
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    I don't see transient suppression caps across the 555 timer. Try using a 0.1uF and a 220uF cap in parallel across the Vcc and GND terminals of the 555.

    When the 555 changes output states, there is a momentary dead short across Vcc and GND; capacitors will help a great deal to quiet that noise down.
     
  7. rbkxiiowe

    Thread Starter Member

    Mar 13, 2009
    15
    0
    Thanks for your help so far everyone.

    I tried the 10uf cap across the gain pins and it had no effect.
    I tried the 0.1uF and a 220uF caps in parallel across the Vcc and GND terminals of the 555 and that had no effect.

    I found I could get rid of the clicks by putting a resistor and capacitor from the output to ground (I assume this is acting as an audio filter? I really am new to this stuff!). This got rid of the clicks but also got rid of the higher frequencies of the guitar audio, which isn't much good...

    I tried to find out about rc filters and found this page. I used a 1uF cap and a 10k resistor in both the high pass and low pass configurations after the square wave and before the LED, but that seemed to kill the whole thing. Is that what Bill_Marsden was suggesting or did I misunderstand?

    This is turning into an essay, but anyway, I took PRS's advice on trying a sine wave so I looked up how to make a simple sine wave oscillator and found this. I used a TL072 instead of the LMH6622 but it seemed to work, and I used a 1M potentiometer for R1 to change the oscillation frequency. This made an LED flash nicely, so I tried the output into the homemade LED+LDR thingy and the same LM386 setup.... and instead of a clicking I get a sort of 'farting' sound in time with the oscillations!

    I don't know enough about this stuff to work out how the clicking/farting noise from the oscillator is getting to my guitar amp's input. Is it getting to it through the ground connection?

    I attach a photo of my breadboard to see if that helps. (The guitar input is on the left and the output to the guitar amp is on the right. The LED is just so I can see the oscillator speed.)

    Cheers for your help so far. This seems to be an excellent forum and it's helping me slowly chip away at the many many bits I don't understand!


    [​IMG]
     
  8. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    You're on the right track. I was talking about soft switching the LM386, which is why I said use it after the 555, between pin 3 of the 555, and the power supplyof the LM386 circuit, so that "click" the 555 was producing didn't go through.

    You're also on the right track with the protoboard, trial and error is the key, IMO.
     
  9. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    No.
    You were told to connect the 10uF cap in series with the gain-setting resistor (the LDR) so that the output of the LM386 does not have its DC level abruptly changed by the 555 signal.

    They are recommended on the datasheet for the 555.

    You need to filter the square-wave to the led, not filter the audio. The square-wave should be a sine-wave.

    The 10k resistor is much too high and the 1uF capacitor is much too low.
    Swap the positions of the 470 ohm resistor and the LED. Then connect a 220uF capacitor across the LED. The negative wire of the capacitor connects to 0V.

    1) Wrong circuit. It is designed to produce and filter high frequencies.
    2) Wrong opamp. The TL072 does not have enough output current to drive the very low value resistors in the filter.

    You can re-design the high frequency high current circuit you found to work, but it is easier to just use a low frequency high current sine-wave circuit instead.

    You forgot to show a schematic of it for us to see what drives it and what limits the current to it.
     
  10. rbkxiiowe

    Thread Starter Member

    Mar 13, 2009
    15
    0
    OK Audioguru. I'll reread that tomorrow and try to work it out. Thanks.
     
  11. RAH1379

    Well-Known Member

    Dec 13, 2005
    69
    1
    I'm using a 555 timer to produce the square wave which then goes into a homemade optoisolator

    pass the square wave through a tank circuit to change it to a sinewave, the squarewave has too many harmonics.

    you might be interested in this site also, www.firebottle.com/ampage it is music electronics, lots of good techs there and builders.
     
    Last edited: Apr 18, 2009
  12. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
  13. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The tremolo circuit from ESP uses a triangle-wave instead of a sine-wave.
     
Loading...