Any complete reference about voltage buffers on the web?

Discussion in 'Electronics Resources' started by cumesoftware, Mar 10, 2008.

  1. cumesoftware

    Thread Starter Senior Member

    Apr 27, 2007
    1,330
    10
    Hello,

    I'm planning to use a voltage follower (also voltage buffer or buffer amplifier) next to the LM1036 audio control, in a project similar to the "Quatour" project in the Projects Collection". Although the LM1036 can drive a power amplifier, it cannot drive headphones. My plan is to use the buffer to drive 16Ohm or 32Ohm headphones, via a TRS socket that automatically switches the sound going to the power amplifiers to the headphones. The power amplifiers I'm planning to use are the TDA2003.

    I could acquire a voltage buffer, but they are expensive and need a split power supply. I'm planning to use a FET, since they don't introduce unpleasant odd harmonics (unlike the bipolar transistors), but not MOSFETS. Is there a way of doing that? I need a reference on voltage buffers made with bipolar transistors or FETs. I already searched on Wikipedia, but there are no details on doing that (on how to calculate resistors, on how to choose the appropriate FET, etc). Is there any complete reference about the subject on the web? I would like to do this by myself, but I need a reference.

    Thanks.
     
  2. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
  3. cumesoftware

    Thread Starter Senior Member

    Apr 27, 2007
    1,330
    10
    Thanks. It might be useful.
     
  4. cumesoftware

    Thread Starter Senior Member

    Apr 27, 2007
    1,330
    10
    I don't understand what means Id, Idq, Vgsq, Vdsq. Also, I don't know the best solution for a unity gain JFET amplifier, if the design against device parameter variations, or the design for midpoint biasing, or the design for zero thermal drift. What is the best? Also, I don't know how to set the gain. I pretend to use an amplifier with unity gain.

    I don't know much about transistors working in the active region as well.
     
  5. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
    I think I led you in the wrong direction. The link I have you was for making a voltage amplifier with a JFET. In your case, you have no need for this, since you wish only current gain and unity voltage. This case would require a common-drain amplifier topology.

    http://en.wikipedia.org/wiki/Common_drain

    Steve
     
  6. cumesoftware

    Thread Starter Senior Member

    Apr 27, 2007
    1,330
    10
    Thanks. So, it is like the common collector configuration topology with a bipolar transistor (often used as voltage regulator boosters), right? But won't this lead to some distortion?

    I'm asking this because with a bipolar transistor the emitter voltage has 0.7V less than the base voltage, and it also depends on the load.

    My application demands loads from 8Ohm (or 16-32Ohm headphones) to 10KOhm, that is, it drives headphones or speakers but it should drive a 10KOhms load represented by the TDA2003 when no speaker is connected.
     
  7. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
    I am under the impression that you'll get distortion either way. You must bias the FET in its linear region and its deviation from the ideal straight line will lead to distortion of your signal.

    This is pretty much what the doctor ordered though, you wanted to use a FET and you want to have high input impedance and low output impedance. You should perhaps breadboard it and check if it is acceptable or not.

    Have you considered running differential configurations? My friend is really into audio design and everything he does is differential.

    Steve
     
  8. cumesoftware

    Thread Starter Senior Member

    Apr 27, 2007
    1,330
    10
    I will give it a try.

    By the way, I found this website:
    http://www.piclist.com/images/ca/ua...notes/node92.html#SECTION00670000000000000000

    What worries me, is the part that says gm * Rs >> 1 for an almost unity gain. I checked that the transconductance gm for a JFET is about 2000uMho, or 0.002Mho. What worries me is if that the units in the formula are expressed in uMho or Mho? If they are in Mho (which is my best guess), I should have at least 100Mho of transconductance, and I can't find such JFET.
     
Loading...