Biasing a TL082 for Single Supply

Discussion in 'General Electronics Chat' started by DanRilley, Nov 18, 2009.

  1. DanRilley

    Thread Starter Senior Member

    Jan 13, 2008
    107
    0
    Hi, I'm trying to recreate the following circuit using a single 6V supply. So far its not working. I've attempted to bias both the TL082s by connecting a voltage divider to the non-inverting inputs, but this didn't work, so I'm wondering if anyone has suggestions.

    [​IMG]

    I realize the schematic calls for +-12V but I figured it still should work at 6V. Let me know if I'm a fool.

    Thanks,
    Dan

    Also as a note, I'm used to working with TL082, but usually the input goes into the inverting input and the non-inverting pin is grounded. So to use it with a single supply, instead of grounding that pin, I put it through a 50/50 voltage divider. However in this circuit the input goes into the non-inverting side so I'm stumped.
     
    Last edited: Nov 18, 2009
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    At the most optimistic, a TL082 will only be able to get within about 1.5v of the supply rails, which would give you about a 3v range in inputs and outputs.

    By specifications, they only need to get within 3v of the supply rails.

    If you're trying to use a TL082 that meets only the minimum specifications with a 6v supply, you have an input range of 3v to 3v and an output range of 3v to 3v; essentially nothing in, nothing out.

    You'll have to increase your supply voltage, or go to an amplifier that has rail-to-rail inputs and outputs.

    Sorry 'bout that - that's just the way it is.
     
  3. DanRilley

    Thread Starter Senior Member

    Jan 13, 2008
    107
    0
    I figured I was asking too much. Anyway the reason I though it 'could' work off 6 volts is because I've built this circuit and it works just fine. Is it because this one is inverting?

    [​IMG]

    Thanks,
    Dan
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    It's likely because one of the TL082's you're using meets the most optimistic specs, and the other one only meets the minimum specs.

    Any time you get "close to the edge" of the specifications, you may get unexpected results.

    That Theremin schematic isn't so wonderful. R8 and R9 are so low as to almost be non-existent.
     
  5. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The TL081 has the Opamp Phase Inversion problem where when one input's voltage is about 3V or less from the negative supply voltage then the output suddenly inverts.
     
  6. peranders

    Well-Known Member

    May 21, 2007
    87
    0
    The supply voltage is too low for this opamp. The whole design can surely be operated on 6 V but you must redesign it a bit. All heavy loads must go to minus and the signal ground must be fed from a virtual ground circuit which can be as simple as a voltage divider (if there are small current involved) or some active solution.
     
  7. DanRilley

    Thread Starter Senior Member

    Jan 13, 2008
    107
    0
    Well thanks for the suggestions.

    I will give reworking the circuit a few hours otherwise I will search for another basic compressor circuit that can run of 6V (I'm kind of constrained there).

    In terms of the Theremin circuit, I'm not using the actual theremin part (the LEDs) so it works fine for my purposes. I just needed a simple way to make a Square Wave sound more like a Sine Wave, this circuit works pretty well. The only problem is that depending on the shape, the amplitude varies a lot so thats why I was interested in adding the compressor.

    Back to the drawing board.

    Thanks again you guys are key.
     
  8. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,534
    Why not give us more information? A triangle wave sound closer to a sine wave, but not exactly like one. A triangle can also be shaped pretty easily to a closer approximation of a sine wave.

    The wave form, is it from an external source, or are you generating it in your project?
     
  9. DanRilley

    Thread Starter Senior Member

    Jan 13, 2008
    107
    0
    Hey, I haven't checked the boards for a while so didn't see this reply.

    I'm generating a square wave with a 74HC14, however I am also capable of generating a triangle with a TL082 (it does work at 6V). The shaping circuit is pretty nice, the only problem is that the more 'sine shape' I give it the lower the amplitude gets. So it would be great to have something that either shaped it without losing amplitude, or, something similar to a compressor, so the amplitude was always the same.

    Again, I have a 6V restriction.
     
  10. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    If you want to not lose the amplitude, you will need to go to a rail-to-rail input and output (RRIO) opamp like an LM6132, MCP6022, etc.
     
  11. DanRilley

    Thread Starter Senior Member

    Jan 13, 2008
    107
    0
    Thanks for the info. These look perfect for my situation, unfortunately they're not supplied by my usual vendor, but I'll keep looking. Thanks again.
     
  12. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    There are many more RRIO opamps than just what I listed; that was only a couple out of hundreds, if not thousands that are being made today.

    The trouble you will have is finding one that is available in a package that you can easily work with (ie: DIP rather than SOT-23 or SOIC, etc.) in the right voltage range with a decent bandwidth.

    Since you have not provided your general location (even just a country would help a great deal) we would have a difficult time recommending an opamp that you could actually purchase where you live.
     
  13. rogs

    Active Member

    Aug 28, 2009
    279
    37

    I use a circuit based on the attached sketch to generate a pretty decent sine wave from a TLO64. I'm guessing the TLO84 would give similar results.
    Using a single 5V supply, you can expect to get about 2V p-p out, before the limitations of the op amp clip the output. Using one of the devices suggested by Sgt Wookie should improve that considerably, I would think.

    I find that using a triangle, rather than a square wave, as a source helps a lot with the effectiveness of the following band pass filter in generating a 'purer' sinewave output from such a simple circuit. Using the resistors around the inverting buffer amp '3' on the sketch, to adjust the gain of the triangle input to the filter, allows you to optimise the output level.

    Filter design --'Q', frequency, etc -- is much simplified if you use something like the freebie Texas Instruments design application 'Filter Pro': http://focus.ti.com/docs/toolsw/folders/print/filterpro.html

    Just some alternative ideas you might like to consider!
     
    Last edited: Nov 21, 2009
  14. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Rogs,
    That isn't going to help him - the TL06x isn't RRIO.
     
  15. rogs

    Active Member

    Aug 28, 2009
    279
    37
    I understand that --I wasn't suggesting using a TLO6x for this project, which is why I suggested the OP look at the one of devices you recommended.

    What I was trying to show was a way of combining the oscillator and filter using one device, rather than using a 74HC14 as the source generator.
    Sorry to confuse -I didn't word things vey well!
     
  16. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You could use a 74HCU04 unbuffered hex inverter connected as a ring oscillator with a half-dozen 1N914/1N4148 diodes & three caps to generate fairly decent 3-phase sine waves, then amplify just one of the three phases using an RRIO opamp. Of course, you'd still need the RRIO opamp anyway...
     
  17. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The MC3317x has a minimum supply of only 3V and a bandwith to 35kHz. It is low power.
    The MC3407x also has a minimum supply of 3V and a bandwidth of 100kHz. It is normal power.

    The TL06x, TL07x and TL08x have a minimum supply of 7V and have the phase inversion problem.
     
  18. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Sure you didn't mean 1.4MHz? 35kHz would make a 741 look like a speed demon...

    And, I think you meant gain bandwidth product of 4.5MHz for these..
     
  19. rogs

    Active Member

    Aug 28, 2009
    279
    37
    I never knew that!
    Taking the information on the data sheet, which shows the max output swing v supply voltage graph extending below +/-2V, I've been happily using TLO64s with a single split 5V supply. Apparently with no problems. Sounds as if I've been lucky so far then!
     
  20. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    You've been "pushing the specs" quite a bit.

    In order to make certain that a given circuit will work, you need to plan for the minimum and maximum ratings. Don't make the assumption that the "typical" specification will work all the time, or you are bound to at some point, be disappointed in the results you get.
     
Loading...