level shifter from +5 to +/- 2.5

Discussion in 'The Projects Forum' started by lavince, Mar 10, 2010.

  1. lavince

    Thread Starter New Member

    Mar 10, 2010
    3
    0
    Hi,
    I am trying to build a +5v to +/-2.5v voltage shifter for my ramp waveform (from MCU). I looked at similar threads but they all seem to be + to + or + to -. What's a good approach to design my specification?
    Thanks!
     
  2. ifixit

    Distinguished Member

    Nov 20, 2008
    639
    108
    You could use an opamp.
    What is the drive power required and the switching speed?
     
  3. lavince

    Thread Starter New Member

    Mar 10, 2010
    3
    0
    The speed I am looking for is 10kHz and I will need to drive <100mW. Since this project is powered by a battery, would a single supply opamp get the job done?
     
  4. ifixit

    Distinguished Member

    Nov 20, 2008
    639
    108
    Two supplies would be needed to power an opamp when the output must go plus and minus wrt ground.

    I have attached a common base PNP transistor configuration that is often used as a discrete solution for level changing. However, it may not have enough drive or accuracy for your needs.



    May not be as trivial as I first thought. More info required.
    • You are driving a 100mW load maximum with no minimum specified. Correct?
    • At ±2.5V that's a 5V swing, so that equates to ±10mA sink/source max. i.e 250Ω resistive load referenced to ground (0V). Correct?
    • What accuracy do you need on the ±2.5V levels?
    • Since the output must go down to -2.5V, where is that bias voltage coming from if there is only one battery?
    The devil is in the details.

    Ifixit
     
  5. lavince

    Thread Starter New Member

    Mar 10, 2010
    3
    0
    sorry for the last response:


    • You are driving a 100mW load maximum with no minimum specified. Correct?
    That's correct.

    • At ±2.5V that's a 5V swing, so that equates to ±10mA sink/source max. i.e 250Ω resistive load referenced to ground (0V). Correct?
    Yes.

    • What accuracy do you need on the ±2.5V levels?
    I actually didn't think of that. I think 5% would be more than sufficient.

    • Since the output must go down to -2.5V, where is that bias voltage coming from if there is only one battery?
    I am thinking of using a DC-DC converter from Maxim which outputs + & -v.

    Sorry for my lack of knowledge. Thanks again for your help!
     
  6. ifixit

    Distinguished Member

    Nov 20, 2008
    639
    108
    Try this circuit, it uses an opamp. You'll need to supply ±5V to power it.

    The LMC6482 has a slew rate of 1 V/uS, so the output square wave has rise and fall times of approximately 5uS. Is that acceptable?

    You can choose your own opamp. Look for one with slew rate better than above if needed; can work with ±5 Volt rails; can drive 10mA into a 250 ohm load.

    Good Luck,
    Ifixit
     
    lavince likes this.
Loading...