+ Post New Thread
Results 1 to 10 of 10
  1. #1
    Full Member level 5
    Points: 2,188, Level: 10

    Join Date
    Jun 2014
    Posts
    273
    Helped
    0 / 0
    Points
    2,188
    Level
    10

    determining the op-amp stability state

    Hi i have attached an image of a very simple op-amp. I am learning the basics of op-amp.
    Click image for larger version. 

Name:	stability.png 
Views:	4 
Size:	48.9 KB 
ID:	130712
    The 1st image we have the opamp amplifying the difference of the two inputs
    and therefore the output voltage, X = 0.99 which is depicted in figure 2. I was taught (just a simple example as such) that now if you take the difference of the two inputs 1V- 0.99V = 0.01. This difference multiplied by gain => 0.01 * 99 which results in an output of 0.99 again. And state will always remain the same and this is known to be the op-amp stable point. I can understand that the op-amp inputs are almost equal which is wat feedback is used for. This example is very simple.


    But say i have 3V in the non-inverting terminal and 2V in the inverting terminal and gain is 4. So at first (3-2) *4 = 4V. So now this 4V output is given to the non-inverting terminal.
    So next it will be (3-4) * 4 = -4. (-4 now to the non inverting terminal) and so
    Next it will be (3-(-4))*4 = 28V. And say out power supply can supply only to 15V So now output will be 15V and will be fed back. And so (3-15)*4 =-48V... This example keeps changing the output. How can i bring it to a stable state like in the first example shown. Like to find the stable point for this example in a intuitive way without any equation. Can someone pls help!!! Thank you!!! :)

    •   AltAdvertisment

        
       

  2. #2
    Advanced Member level 5
    Points: 18,118, Level: 32
    Achievements:
    7 years registered
    crutschow's Avatar
    Join Date
    Feb 2012
    Location
    L.A. USA Zulu -8
    Posts
    3,527
    Helped
    831 / 831
    Points
    18,118
    Level
    32

    Re: determining the op-amp stability state

    From your first example you state the two op amp inputs are almost equal at the stable point, so just apply that reasoning to the second case.

    When 3V is applied to the non-inverting input, the output will start to increase until the inverting input voltage essentially equals the non-inverting input voltage. Thus the output voltage equals the input voltage at the stable point.

    Don't know where you came up with a gain of 4 for the op amp.
    Op amps typically have an open-loop gain of over a 100,000.

    And how does 4V from the output get to the non-inverting terminal? It goes to the inverting terminal.
    The non-inverting terminal is fixed at the 5V input voltage.
    Zapper
    Curmudgeon Elektroniker



  3. #3
    Super Moderator
    Points: 64,658, Level: 62
    Achievements:
    7 years registered
    Awards:
    Most Frequent Poster 3rd Helpful Member

    Join Date
    Apr 2014
    Posts
    13,131
    Helped
    3031 / 3031
    Points
    64,658
    Level
    62

    Re: determining the op-amp stability state

    Hi,

    But say i have 3V in the non-inverting terminal and 2V in the inverting terminal and gain is 4. So at first (3-2) *4 = 4V. So now this 4V output is given to the non-inverting terminal.
    So next it will be (3-4) * 4 = -4. (-4 now to the non inverting terminal) and so
    Next it will be (3-(-4))*4 = 28V. And say out power supply can supply only to 15V So now output will be 15V and will be fed back. And so (3-15)*4 =-48V... This example keeps changing the output. How can i bring it to a stable state like in the first example shown.
    What you describe is the problem with digital regulation loops.
    You are thinking in "steps". Steps in time and steps in voltage.

    But this is not true with analog systems.
    When IN+ = 3V, IN- = 2V and gain = 4 then you imagine the output jumps from 2V (because output = IN-) to 4V.
    But it doesn´t jump. It moves. Every OPAMP has a limited slew rate. For your example let´s assume it is 1V/us.

    so imagine 10 ns of time later..
    (now you may say: this 10ns is a step, too! True. But while the voltage moves continously, not in steps, we just take a snapshot 10ns later of the voltage to see what happens. You may vary the timng of your snapshots... but the result will be the same..)

    Within the 10ns and a slew rate of 1V/us the voltage moves 0.01V

    Now the voltages are 3V, 2.01V --> (tries to target) 3.96V
    10ns later: 3V, 2.02V --> 3.92V
    10ns later: 3V, 2.03V --> 3.88V
    and so on...
    after a while:3V, 2.38V --> 2.48V
    10ns later: 3V, 2.39V --> 2.44V
    10ns later: 3V, 2.40V --> 2.40V (hit the target. Stable)

    (This is a simplified view!)

    Klaus


    2 members found this post helpful.

  4. #4
    Super Moderator
    Points: 247,151, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    42,972
    Helped
    13075 / 13075
    Points
    247,151
    Level
    100

    Re: determining the op-amp stability state

    The original post is showing that a negative feedback loop with pure delay (your assumed step delay) but no bandwidth limitation won't be stable.

    Any real amplifier and respectively any feedback loop has however bandwidth limitations (poles in terms of linear system analysis). A loop ruled by a single dominant pole (the case of typical unity gain compensated OP) will be generally stable.

    A negative feedback with delay plus slew rate limitation (a non-linear system, by the way) will show limit cycles and is not stable.



    •   AltAdvertisment

        
       

  5. #5
    Advanced Member level 5
    Points: 39,988, Level: 48
    Achievements:
    7 years registered
    LvW's Avatar
    Join Date
    May 2008
    Location
    Germany
    Posts
    5,733
    Helped
    1718 / 1718
    Points
    39,988
    Level
    48

    Re: determining the op-amp stability state

    Quote Originally Posted by FvM View Post
    A negative feedback with delay plus slew rate limitation (a non-linear system, by the way) will show limit cycles and is not stable.
    This sounds as if you are saying "always" ! ?



  6. #6
    Super Moderator
    Points: 247,151, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    42,972
    Helped
    13075 / 13075
    Points
    247,151
    Level
    100

    Re: determining the op-amp stability state

    I presume pure delay plus slew rate limitation without a pole is unstable. But it's a theoretical case.



  7. #7
    Full Member level 5
    Points: 2,188, Level: 10

    Join Date
    Jun 2014
    Posts
    273
    Helped
    0 / 0
    Points
    2,188
    Level
    10

    Re: determining the op-amp stability state

    Thank you all for the reply!!! I knew that opamp gain is generally very high. But i found the gain mentioned to be 4 on the following link
    http://www.sanfoundry.com/linear-int...configuration/
    and i'm sorry i mis-typed it. Yes the 4V goes to the inverting terminal and the non-inverting is fixed.
    Last edited by preethi19; 15th July 2016 at 12:50.



  8. #8
    Advanced Member level 5
    Points: 39,988, Level: 48
    Achievements:
    7 years registered
    LvW's Avatar
    Join Date
    May 2008
    Location
    Germany
    Posts
    5,733
    Helped
    1718 / 1718
    Points
    39,988
    Level
    48

    Re: determining the op-amp stability state

    Quote Originally Posted by preethi19 View Post
    Thank you all for the reply!!! I knew that opamp gain is generally very high. But i found the gain mentioned to be 4 on the following link
    http://www.sanfoundry.com/linear-int...configuration/
    Yes - the contents of the link is somewhat confusing.
    In the headline, they speak about "open-loop" - and they show an ampfier symbol (triangle) with a gain of 4.
    Certainly, this symbolizes an opamp with a negative feedback and a closed-loop gain of 4 (but they should mention it).

    My recommendation: Never blindly trust any information found in the internet.



  9. #9
    Advanced Member level 5
    Points: 14,377, Level: 28

    Join Date
    Nov 2012
    Posts
    2,660
    Helped
    619 / 619
    Points
    14,377
    Level
    28

    Re: determining the op-amp stability state

    Quote Originally Posted by preethi19 View Post
    Hi i have attached an image of a very simple op-amp. I am learning the basics of op-amp.
    Permit me to remind the very basics:

    1. An ideal op-amp has infinite gain. This number does not play in the gain of may practical circuit.

    2. The two inputs (inverting and non-inverting) are always at virtual ground (what is that!)

    3. All practical circuits using op-amp used as an amplifier use feedback.

    4. The gain of an op-amp with feedback depends on the resistors in the network- and not on the actual gain (which is anyway very large) of the op-amp.



    •   AltAdvertisment

        
       

  10. #10
    Advanced Member level 5
    Points: 39,988, Level: 48
    Achievements:
    7 years registered
    LvW's Avatar
    Join Date
    May 2008
    Location
    Germany
    Posts
    5,733
    Helped
    1718 / 1718
    Points
    39,988
    Level
    48

    Re: determining the op-amp stability state

    Quote Originally Posted by c_mitra View Post
    Permit me to remind the very basics:
    2. The two inputs (inverting and non-inverting) are always at virtual ground (what is that!)
    Just a small correction:
    * for inverting circuits, the non-inv. input is at real (physical) ground and the inv. input is at virtual ground
    * for non-inv. circuits none of the inputs is at virtual ground.



--[[ ]]--