Problems with the basic LM386 circuit.

Discussion in 'The Projects Forum' started by Gremlinbrd, Jul 16, 2006.

  1. Gremlinbrd

    Thread Starter New Member

    Jul 16, 2006
    6
    0
    I have put together the LM386 circuit just as described in the national semiconductor typical application page. The problem I’m having is that I get a loud buzz out of the speaker. I have tried the different circuits listed on the page, replaced the IC and the POT, and caps but still get a terrible buzz. I am using a 9v battery to power. One thing I have noticed is when I touch the pot to adjust, the buzz varies also when I touch the speaker’s metal frame. I have google the hell out of this and could not find a similar problem. This shouldn’t be so tough!
     
  2. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    If you will identify which circuit in the application note 263 you are using maybe we can help you figure out what is going on.

    hgmjr
     
  3. Gremlinbrd

    Thread Starter New Member

    Jul 16, 2006
    6
    0
    lm386.pdf

    I have tryed the three circuits on page 5, Amplifier with gain of 20, 200, and 50.

    thanks
     
  4. JoeJester

    AAC Fanatic!

    Apr 26, 2005
    3,270
    1,064
    Are you talking a low frequency buzz? Say ... 120 Hz?
     
  5. Gremlinbrd

    Thread Starter New Member

    Jul 16, 2006
    6
    0
    It sounds lower than 1k yet it varies with the position of the pot. It sounds like someone is touching the input, like RF interference but much louder.
     
  6. JoeJester

    AAC Fanatic!

    Apr 26, 2005
    3,270
    1,064
    Do you have an oscilloscope to measure the buzz ?

    Is it present when there is no input connected?
     
  7. Gremlinbrd

    Thread Starter New Member

    Jul 16, 2006
    6
    0
    The buzz is present with or without an input signal. When I do have an input (my iPod mini) I can hear the music at certain positions of the pot (low and distorted), but most of the way through the turn it is just loud noise.

    Yes I have a scope. I will have to take a look at that tomorrow.
     
  8. Gadget

    Distinguished Member

    Jan 10, 2006
    613
    0
    Sounds like an earthing problem. Check the pot is hooked up correctly, check all earths are going to a single point, check there are no earth loops, make sure the input/pot earth to central earth point is independent of power supply earth up to the central earth point, and the speaker earth up to the central earth point.
    Essentually, star all the earths off from a single point on the PCB, making sure there are no extra earth paths.
    Also try earthing the body of the pot (if metal).
     
  9. Gremlinbrd

    Thread Starter New Member

    Jul 16, 2006
    6
    0
    With the scope I found the noise was around 70-80Hz at a low level and around 3kHz at a higher volume. How do I make the input/pot ground independent of the 9v ground?
     
  10. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    After looking at the datasheet that you attached in one of your earlier posts, I noticed that the schematics of the amplifiers you are trying to construct do not show any power supply bypass capacitors.

    I would suggest, if you haven't already done so, that you connect a 0.1 microfarad capacitor between the power pin of the LM386 and the ground pin of the device. I would then get a polarized capacitor (electrolytic or tantalum) in the neighborhood of 10 microfarads or higher rated for 12 volts or higher and put it across the power and ground pins of the LM386. Be sure that you connect these capacitors as close to the power and ground pins of the LM386 as possible and keep their leads as short as practical.

    I suspect that the noise you are experiencing is due to a lack of power supply filtering.

    hgmjr
     
    Treeman and CaptainPrice like this.
  11. JoeJester

    AAC Fanatic!

    Apr 26, 2005
    3,270
    1,064
    Check and ensure you follow what gadget suggested also.

    This seems like a loose ground or too much ripple on the power source type problem.
     
  12. Gremlinbrd

    Thread Starter New Member

    Jul 16, 2006
    6
    0
    hgmjr,

    Please see attached jpg to see if I understood you correctly. I am building this to be used as a utility monitor (a cheap version of the Whirlwind Q-Box http://www.whirlwindusa.com/test.html). Should I then put the .1 micro F cap on the chassis of the XLR and 1/4” input?

    One reason I do not believe it is a ground loop or a loose ground, is that I have two versions of this circuit built. I have one built soldered on copper clad board and another I have built on a bread board to quikly try new configurations, and both make the same noise.

    Thanks everyone for your input thus far
     
  13. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    I looked at your schematic attachment. You have connected the 10 microfarad cap exactly where I suggested. All you need to do is connect the 0.1 microfarad in parallel with the 10 microfarad cap. I know it sounds a bit goofy to put a 0.1 microfarad cap in parallel with a 10 microfarad cap but there is reason for it. The 0.1 microfarads cap is better suited to handling the high frequency power supply noise while the 10 microfarad cap works at the low frequency range. By placing the two caps in parallel you get filter coverage over a broader range of frequencies.

    I can't promise that this will cure your problem but it is certain to yield a more robust design.

    hgmjr
     
  14. JoeJester

    AAC Fanatic!

    Apr 26, 2005
    3,270
    1,064
    your last attachment doesn't show up at all.
     
  15. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    JoeJester,

    I too had difficulty opening the attachment. I finally ended up downloading the file to my computer and then I was able to view it.

    hgmjr
     
  16. mozikluv

    AAC Fanatic!

    Jan 22, 2004
    1,437
    1
    hi hgmjr,

    that is a valid observation on the national semi suggested circuit in using their device. not all of their audio circuits have that filter shown. i believe that they have assumed that whoever will build their suggested circuit knows the proper power supply line to the device filtering technique.

    if by employing the supply filter and the buzz is still there, it is highly probable that the chip is defective or a fake chip. a lot of that particular chip are fakes proliferating here in asia.

    moz
     
Loading...