Audio signal adder for headphones

Discussion in 'The Projects Forum' started by raisintoe, Mar 7, 2016.

  1. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    I am creating an audio signal adder using this op-amp. I do not need to amplify the audio.

    I want to drive my headphones directly from the op-amp, but I do not know how to tell if this op-amp can handle the load of a typical headphone set, (there is no power dissipation rating or anything like that in the op-amp datasheet).

    If I should not try a direct drive to my headphones, how can I add a simple transistor circuit to drive my headphones?
     
  2. crutschow

    Expert

    Mar 14, 2008
    13,049
    3,244
    If you look on page 6 of the data sheet you referenced you will see that the smallest load it is rated to drive is 600Ω so, if your headphones have a lower impedance than that (and they likely do) then you will need an added amplifier.
    The easiest is to just use a IC audio amp such as one of these.
     
  3. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    OK, I would rather make an amplifier using an npn transistor, since I have those on hand. I guess what I want is a common emitter amplifier, but I still have trouble understanding how they work.

    Here is an example, I have trouble getting this to work properly on LtSpice.
     
  4. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    Or, since I only need to amplify current, a common collector amplifier is probably better.
     
  5. crutschow

    Expert

    Mar 14, 2008
    13,049
    3,244
    Here's a discussion of simple amplifiers and a headphone amp using two NPN transistors (Question 19).
     
  6. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,440
    368
  7. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,546
    1,252
    If you don't have any PNP transistors, then an all NPN amplifier will not be very efficient. What is your intended power source?

    ak
     
  8. ScottWang

    Moderator

    Aug 23, 2012
    4,855
    767
    An easy method you may try it, using a class A amplifier as the first circuit in your linked page, and adding a bjt Totem-Pole to amplifier the signal.
     
  9. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    I will set up a DC power supply under 12 volts. So I can choose anything under 12V DC

    Also, I might have a few PNP transistors hanging around, that may be a possibility.
     
    Last edited: Mar 7, 2016
  10. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    Here is a schematic I drew up, using the amplifier circuit that Crutschow found.

    The 32 ohm resistor at the right is my headphones

    Schematic.png
     
  11. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,546
    1,252
    The idea is ok, but the implimentation needs a little work...

    Even though you have bipolar supplies, the output stage swings only between P and GND. This limits the output power to 1/4 its possible maximum value.

    The opamp has a totem pole output stage, meaning it can both source and sink current. For this reason, I don't think you need R8 as a pull down.

    R7 can be smaller to increase the output voltage swing.

    If you search for opamp booster schematic, you will find many examples of ways to boost the opamp output current while maximizing output voltage swing. Some are so efficient that you would get more output with only one power rail than your circuit gets with two.

    ak
     
  12. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    This is the exact circuit that I have on my breadboard. It works pretty good, but sounds muddy, It's good for a first try! I think the low quality is mostly due to R14, I gave it a large resistance to avoid pulling too much current through it. What is a good way around this problem? Breadboard circuit.png
     
  13. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,546
    1,252
    By the time you have R14 down low enough that the circuit almost has a symmetrical output impedance, Q3 is burning up due to the static current. Congratulations - you've just re-discovered the problem with Class A audio circuits - they suck - - up lots of power.

    The only way around the limitations of a single emitter follower circuit is not to use an emitter follower circuit. If you search for things like '1 watt audio amplifier schematic' or 'opamp output booster', you will get hundreds of circuits to choose from. My favorite is the one with one NPN and one PNP in the power leads of the opamp, but most people prefer a complimentary emitter follower on the opamp output.

    Plus there a threads on this site that go into this.

    ak
     
    Last edited: Mar 15, 2016
  14. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    Yes, I did search for opamp booster. I have seen the simple PNP NPN setup, but how do I keep this circuit from clipping between 0.7 and -0.7 volts? The signal from the opamp has to pass the threshold of the transistors before they respond.
     
  15. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,546
    1,252
    Depending on the exact configuration, the opamp supplies some of the output current in this voltage range, plus the negative feedback loop shifts the output stage operating point to supply the rest. Remember, when the two emitters are making +1 V (for example) the output of the opamp is at +1.6 V. If you post a schematic, we can discuss it. No schematic, no details.

    ak
     
    raisintoe likes this.
  16. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    This is all I am talking about

    Opamp booster.png
     
  17. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,546
    1,252
    Q2 is upside-down.

    Disconnect the right side of R1 from the op37 and connect it to the two emitters. This brings the two transistor base-emitter junctions inside the feedback loop, where the opamp can compensate for them. to reduce crossover distortion, add a resistor from the two bases to the two emitters so the opamp can provide some current to the output directly.

    ak
     
    raisintoe likes this.
  18. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    Awesome! I made the changes, and will put this circuit together today. How does the added resistor reduce distortion? Or, how does distortion occur during crossover? Is it simply the inrush of current when the transistor is turned on that causes distortion?
     
    Last edited: Mar 16, 2016
  19. raisintoe

    Thread Starter Member

    Mar 24, 2012
    46
    0
    I've built the circuit, it works really well! I placed a 1K resistor between the emitters and bases, there is still some distortion, I am not sure how much current I can draw through the opamp.
     
  20. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,546
    1,252
    It's on the datasheet.

    ak
     
Loading...