Opamp circuit not yielding expected results

Discussion in 'The Projects Forum' started by Lilrayray, Feb 21, 2009.

  1. Lilrayray

    Thread Starter Active Member

    Feb 21, 2009
    31
    0
    I all, I am working on a small personal project that involves constructing a simple microphone preamp. The microphones I am using have a impedance rating of 680 ohms, and I am attempting to amplify their milivolt output to ~2V, which will later be fed into an ADC.

    I have been doing a considerable amount of reading about opamps (very new to circuit design in general) and have begun tackling the circuit design. Unfortunately, my first attempts have not yielded the results I have expected: the 2mV input signal is not being amplified by even half of the 1000 gain inverting feedback loop.

    I have a feeling that I am missing some fundamental concept about opamps, however, even after reading numerous internet resources and "Handbook of Operational Amplifier Application" (published by TI) I cant seem to figure out what is wrong with my circuit. I have attached both a screen capture of the circuit schematic and of the oscilloscope output and function generator settings (I used multisim: analog devices edition).

    Any help or info on this is greatly appreciated.

    p.s. could anyone recommend a good general purpose single-supply (5v) opamp that would suit non-intensive audio-preamp needs? The one I am currently simulating with is the OP490 from Analog Devices (something tells me this is not intended for general purpose/audio): http://www.analog.com/en/other/militaryaerospace/op490/products/product.html
     
  2. mik3

    Senior Member

    Feb 4, 2008
    4,846
    63
    Use two 1K resistors to make a voltage divider and then connect the non inverting input between them. Also, put a 22uF capacitor on the left side of R1.
    The microphone will be connected between the capacitor and the non-inverting input.
     
  3. Lilrayray

    Thread Starter Active Member

    Feb 21, 2009
    31
    0
    Hmm, Im not sure I quite understand...

    Is this a completely new non-inverting setup, or is this in addition to what I already have?

    If it is an entirely new setup, what about the feedback loop to determine the gain? Also, why would a non-inverting setup alter anything but the sign of the output voltage?

    Also, I have a 10uF capacitor in my mic circuit... I was under the impression that the value didnt matter too much, as it was merely for eliminating the small DC current powering the microphone.

    Any explanation pertaining to this recommended circuit is greatly appreciated!
     
  4. Cabwood

    Member

    Feb 8, 2009
    20
    0
    The first problem you have is that the microphone produces positive-going voltages that your inverting amplifier will try to invert. For instance if the input is +1mV, your amplifier will try to produce -1V output. It can't though, because it doesn't have a negative supply rail to allow the output to swing negative. Your design's output cannot be less than 0V. In practice the output won't swing even as low as 0V.

    The second issue is the input offset voltage of the amplifier. With a closed loop gain of -1000, your circuit will multiply any input voltage offset (which could be of the order of millivolts, depending on the device) by this amount, which may saturate the output even without an input.

    The third issue is bandwidth. You are asking an awful lot of the opamp, by configuring it with a gain of 1000. Some opamps will be fine, others will not have sufficient bandwidth to achieve this. It is better to cascade two amps, each with a gain of 32, to achieve a total gain of about 1000. This way all audio frequencies, right up to 20kHz, will be amplified equally. I assume that this is what you want.

    There are many ways to overcome these problems, but first you have to understand the problems themselves. Questions? Just ask.
     
  5. Lilrayray

    Thread Starter Active Member

    Feb 21, 2009
    31
    0
    Ahh, ok, there is the first thing I misunderstood. I had thought using a single-supply opamp would merely center the amplified signal between 0V and the power source voltage.

    From the sound of things, I am asking too much in terms of gain. How do you ensure that my signal is amplifies ~2-3V? I feel that I will be losing a lot of potential information if I do not generate as much of a voltage as the ADC will allow.

    If i use a bipolar power supply, what would be the easiest way to center the output from the opamps between 0V and +5V? I assume it could be done with another opamp, however it would seem rather inefficient to use 3 opamps for a single microphone's preamp!

    Anyway, thanks for the help!
     
  6. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,535
    Just to add my 2¢, you can do this two ways, either make a voltage divider, which puts the op amp in the linear region, or add another power supply, which is simpler in terms of the op amp circuit, as shown below.

    [​IMG]

    Figure R4 and R5 would be 470Ω, to split the power supply. Op amps in general are happier with dual power supplies. You will need pretty large capacitors for the single version.

    I went ahead and changed it to the non inverting configuration, it doesn't really make much difference either way though.

    I wrote an article on how to make pseudo dual power supplies from a single here.

    PS: I was writing and drawing this at the same time you were adding your reply.

    .
     
  7. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    R1 should be capacitor-coupled to ground so that the opamp does not amplify its DC input offset voltage.

    Is the microphone a dynamic type (coil and magnet) or is it an electret type (condenser with built-in high voltage charge)? An electret type has an impedance much higher than only 680 ohms and it needs a bias current.

    The OP490 quad opamp is very low power which means it has very poor high frequency response. With a gain of 1000 its high frequency response starts dropping above 20Hz.
    I forgot to see if it has low noise.

    If you use an OP4134 low noise quad opamp that has a normal supply current and normal bandwidth then its high frequency response starts dropping above 10kHz when its gain is as high as 1000.
     
  8. Lilrayray

    Thread Starter Active Member

    Feb 21, 2009
    31
    0
    The microphones I will be using are these electret condensers: http://parts.digikey.com/1/parts/1517486-mic-condenser-elect-uni-47db-cmi-5247tf-k.html
    The microphone is listed as having a 680 ohm impedance.

    I wont be using the OP490 (which I really picked semi-randomly from the listed opamps in multisim), as I have decided to stick with a dual supply. Ultimately, I will probably use a few of these: http://focus.ti.com/docs/prod/folders/print/ne5534.html
    Also, I think Ill try cascading two opamps as previously recommended.

    I feel as if I very recently read about it, but how do you determine the actual output impedance of the opamp to determine the resistor values in the second opamp's feedback loop? The "ideal" opamp has an impedance of zero, so I am going to assume in reality the output impedance is relatively small?

    Also, it seems that most have recommended some variation of a non-inverting configuration... does this really have any impact on anything, or is it simply a matter of preference?
     
  9. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,535
    It doesn't really matter. Coupling the input with a cap was good advice though.

    [​IMG]
     
    Last edited: Feb 22, 2009
Loading...