Connect with us

OPA in mic preamp

Discussion in 'Electronics Homework Help' started by ussus, Nov 13, 2016.

Scroll to continue with content
  1. ussus

    ussus

    6
    0
    Nov 12, 2016
    Hello!
    Could you help me to find what's wrong with my preamp buil on the base of OPAMP?
    The amp works in general but after few minutes it starts to raise and produces buzzles and creaks.
    Is this a problem with filtering?
    It works with two condenser cartriges. Schematic below:
     

    Attached Files:

    Last edited: Nov 13, 2016
  2. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    R2/c3 are your problem. The lower end of r2 needs to go to a 0V rail. You need to split your power supply to be +/- 4.5V, or perhaps more easily, change it to +/- 9v (using 2 9V batteries).

    Take a look at typical audio amplifiers using op amps and see how it's done.
     
  3. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    Also, many of the component designations are duplicated.

    You could add 100 to one channel and 200 to the other so that instead of having two R4's, you have an R104 and an R204.
     
  4. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    And thirdly, what is "OPA"?
     
  5. ussus

    ussus

    6
    0
    Nov 12, 2016
    Thanks for quick reply!
    OPA = OPAMP - excuse me for wrong shortening.
    I do not want to use 2 batteries, it has to be as lightweight as possible. One "heavy" battery is more than enough.
    There are many application for OPAMPS to use single voltage, I wish to use this method. If possible...
     
  6. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    Change R2 to 22k, remove C4 (R2 goes straight to ground). As another 22k resistor from +9V to the top of R2.

    If you want to you can connect the unused C4 between +9V and gnd.
     
  7. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    Yes, but not audio ones. You need a mid rail reference point or your op amp won't be able to amplify both halves of the audio signal.

    In this case I've given you a voltage divider with an impedance similar to the resistor that should be connected to a stiff rail. This will work, at the expense of additional current drain from the battery of about 200uA per channel, or less than 0.5mA in total.

    If you were really after a light weight device, you could use two of those small 6V batteries about half the size of a AAA cell.

    Combined with a low current opamp and better design (I.e. Design for a double ended power supply) you would probably get a near equivalent service life with a significant weight reduction.

    Are you using surface mount components?
     
    davenn likes this.
  8. ussus

    ussus

    6
    0
    Nov 12, 2016
    Thanks, it is good idea to find two 6V batteries indeed and rebuild a circuit. :)
    Let you know soon.
     
  9. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    The battery you're looking for is an A544. It's a little larger than I recalled (I was thinking of the thinner 12V battery for remote controls). Actually, here is the smaller one. The A544 looks like a reasonable compromise -- it will have more capacity than the smaller battery.

    Some obvious changes in the circuit will be:
    • Label +9V as +6V
    • Label gnd as -6V
    • New 0V from between batteries in series
    • R5 changing to 15K
    • R3/R4 removed and replaced with a single 33K resistor to 0V
    • R1/R2 replaced with 68K with lower end of R2 connected to 0V
    • R6 and R7 connect to 0V rather than -6V
     
  10. Audioguru

    Audioguru

    3,240
    702
    Sep 24, 2016
    The 5532 and many other opamps work perfectly in that circuit that uses a single positive supply (not a dual polarity supply), is biased properly at half the supply voltage and has the input, output and feedback resistor to ground capacitor-coupled. I noticed that the values for the coupling capacitors are way too large so that they respond to earthquake frequencies. I also noticed that the gain is only 2 instead of from 100 to 500 so a microphone will not work.

    If the circuit is modified to have much more gain then the 10k resistors that power the electret mics need an RC filtered supply voltage.
     
  11. ussus

    ussus

    6
    0
    Nov 12, 2016
    Hi Audioguru.
    At the moment I do not need more gain - mic is working OK, sensitivity is enough. First I wanted to use a potentiometer but decided to use smaller single resistor instead (R1).
    Do you mean to decrease the value of C2? How much?
     
  12. Audioguru

    Audioguru

    3,240
    702
    Sep 24, 2016
    The lowest audio frequency is 20Hz but many speakers cannot go that low. We feel lower frequencies as vibrations. The formula for a capacitor that feeds a resistance is 1 divided by 2 x pi x RC so the frequency where the response begins to reduce for your 10uf capacitor feeding the 10k resistor is 1.6Hz which is much too low.
    What value should C2 be? Use the formula and see that it should be 185nF. but since you have three RC's cutting low frequencies then all three capacitors should have values 3 times the calculated values.
     
  13. ussus

    ussus

    6
    0
    Nov 12, 2016
    I wonder what is the role of these capacitors onto raising of OPAMP? C1 and C2 separates input and output. And in many applications they are much larger than mine.
    There are many samples of audio amplifiers (mic preamps also) which uses single voltage supply, eg.:
    LINK1
    LINK2

    I do not understand the mechanism of raising of this circuit. I want to find what is the source of this effect.
     
    Last edited: Nov 14, 2016
  14. Audioguru

    Audioguru

    3,240
    702
    Sep 24, 2016
    I showed you the formula for calculating the value of the coupling capacitors. The capacitors pass AC but block DC. The value of the capacitor depends on the low cutoff frequency you want and the resistance. For example for a 30Hz cutoff frequency into an 8 ohm speaker the coupling capacitor value must be 670uF which of course is very large. I do not know what you mean when you say "raising".

    I did not see the links yet since I am in the middle of this reply.

    I do not know what you mean when you say "raising". Raising is increasing the height of something.
     
  15. davenn

    davenn Moderator

    13,808
    1,943
    Sep 5, 2009
    you used the term raising in your first post as well
    I have no idea what you are referring to ??? what do you mean by raising ?


    Dave
     
  16. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    I don't understand what "the method of raising" means. Autocorrection failure?

    You're correct that the op amp has no idea whether the supply is single ended our not. My point was that you need a mid rail reference. You can do that with a pair of resistors, but a double ended power supply is an easy way to do that without unnecessary current drain.

    The odd feedback loop you had was designed to produce low frequency roll off in gain. I had assumed you were supposed to have flat gain.

    I can now see what the circuit is supposed to do. C3 provides a great way to inject noise into the circuit though. There are some discussions on this point on the net with the cure bring a low impedance (stiff) mid rail reference, and here we are on that merry go round...
     
  17. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    Triple snap
     
    davenn likes this.
  18. Audioguru

    Audioguru

    3,240
    702
    Sep 24, 2016
    The circuit is a standard amplifier using the opamp with a single polarity supply. It already has a mid rail reference. I have built hundreds of them but I always filter the voltage source of the mid rail reference voltage to prevent the opamp from amplifying noise on the main supply voltage.
    C3 does not inject noise into the circuit because it connects to ground. C3 allows a DC gain of 1.
    The mid rail reference voltage does not need to be "stiff" because all it does is supply the very low bias current to the input of the opamp. But very high resistor values are noisy resistors.
     
  19. Audioguru

    Audioguru

    3,240
    702
    Sep 24, 2016
    I wonder if "raising" is a bad translation of acoustical feedback howling or squealing? It is simple to fix it, turn down the volume or separate the mic from the speaker.
     
  20. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,489
    2,830
    Jan 21, 2010
    If there is any noise on the power rail, the reference voltage will move up and down. If both the feedback loop and the signal are referenced against this, the CMMR of the op amp will take care of the noise. In this case, the two have separate reference voltages. One will vary against the other leading to the noise being amplified.

    A stiff(er) reference voltage that both the input and the feedback loop reference will fix this.

    Where you see the noise being injected is less important than whether it results in a common mode or differential signal.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Similar Threads
Loading...
Electronics Point Logo
Continue to site
Quote of the day

-