Dynamic Compensator - Noise Analysis in OrCAD Capture 16.2

Discussion in 'The Projects Forum' started by pdunc, Feb 26, 2010.

  1. pdunc

    Thread Starter New Member

    Feb 26, 2010
    8
    0
    Good morning!

    I have designed a filter which can reduce dynamic error in measurement systems which exhibit first-order behaviour. I have proven the ideal case using PSpice simulation, now I would like to test the circuits response to noisy input.

    From google it would appear that using the AC Sweep/Noise simulation and enabling "Noise Analysis" should do the job. I have attempted to do this but the simulation results are identical to the ideal case.

    I am unsure if I have set this up correctly, the link below is a screenshot of the circuit and simulation profile (I won't embed as it's large);

    http://i496.photobucket.com/albums/rr326/pdunc/first_order_comp-1.jpg

    Does anyone have experience of testing a schematic with noise? I am expecting to get some noise amplification through the system.

    Thanks in advance,

    Peter
     
  2. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The LM324 quad opamp is very noisy, has a poor high frequency response and has severe crossover distortion.
    Your circuit has low frequencies cut twice and low frequencies boosted one time. Why?
    It has two inverters and a useless opamp so maybe a dual opamp can be used.
     
  3. pdunc

    Thread Starter New Member

    Feb 26, 2010
    8
    0
    Thank you for your response, Audioguru.

    These are indeed considerations I will need to take into account for the final design.

    To explain what things are doing at present...

    -The initial RC circuit is to simulate my input signal; which will be a voltage from a temperature transducer with first-order behaviour.

    -The unity-gain buffer here is self explanatory.

    -The next filter is used to 'extend' the frequency response of the input signal, meaning when the two signals are convolved we should see a faster response time. A by-product of this is signal inversion.

    -The unity gain buffer here is pointless and will be removed.

    -The final op-amp simply inverts back to the correct polarity.

    Hopefully that gives a better idea of what I am trying to achieve. Although none of the above helps with my current problem of simulating a circuit with noise in PSpice.

    Do you have any experience of this?

    Thanks again,

    Peter
     
  4. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    I use low noise TL071 single, TL072 dual and TL074 quad low noise very low distortion audio opamps.
    I saw a very noisy distorted LM324 in a speakerphone that was made in Korea and sounded awful.

    I don't think PSpice knows anything about noise.
     
    pdunc likes this.
  5. pdunc

    Thread Starter New Member

    Feb 26, 2010
    8
    0
    I had built a test circuit with the LM324 and this had poor results - the characteristics were correct, but a number of errors were introduced.

    This is project work for university so the errors are good for discussion points, but I have ordered in a TL074 to try to gain accuracy for the final built.

    Thanks again,

    Peter
     
Loading...