+ Post New Thread
Results 1 to 4 of 4
  1. #1
    Newbie level 5
    Points: 139, Level: 1

    Join Date
    Sep 2019
    Posts
    9
    Helped
    0 / 0
    Points
    139
    Level
    1

    RHP poles always appear in my designed amplifier

    I am trying to use feedfoward compensation in multi-stage OTA for driving large capacitive load. However, it always oscillate, and PZ analysis in cadence spectre always shows RHP pole in my amplifier..

    The circuit I have tried are all from journals. Here is one example https://ieeexplore.ieee.org/stamp/st...number=7389422

    According to the journal's simulation it should not have any RHP pole.
    Does anyone have the same experience?

    Here is my amplifier
    Click image for larger version. 

Name:	amplifier.jpg 
Views:	9 
Size:	147.7 KB 
ID:	157785

    Here is my bode plot of amplifier
    Click image for larger version. 

Name:	bode.jpg 
Views:	6 
Size:	185.3 KB 
ID:	157783

    Here is its poles and zeros
    Click image for larger version. 

Name:	PZ.PNG 
Views:	6 
Size:	131.3 KB 
ID:	157784
    Last edited by kokokosini123; 16th February 2020 at 04:02.

    •   AltAdvertisement

        
       

  2. #2
    Advanced Member level 3
    Points: 6,151, Level: 18
    Dominik Przyborowski's Avatar
    Join Date
    Jun 2013
    Location
    Norway
    Posts
    788
    Helped
    345 / 345
    Points
    6,151
    Level
    18

    Re: RHP poles always appear in my designed amplifier

    I have tried some small gain multistage and they are working on simulations.

    RHP is a sign that somewhere you have positive feedback. It might be result of odd number of stage and tricky connection somewhere.
    Your PZ result showing low freq pole given by high gain Miller, 7th order pole at tens of MHz and RHP. Such opamp (other designs done such times, too) is working with high load (nF order) to ensure stability, this higher order poles caused by diode connected guys in small gain enhancers should be (N+3) times higher than UGF.

    My proposal for debugging. Start with ordinary Miller, than add stage by stage and check in which configuration RHP appears.


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  3. #3
    Super Moderator
    Points: 265,480, Level: 100
    Awards:
    1st Helpful Member

    Join Date
    Jan 2008
    Location
    Bochum, Germany
    Posts
    46,390
    Helped
    14118 / 14118
    Points
    265,480
    Level
    100

    Re: RHP poles always appear in my designed amplifier

    A RHP pole in the open loop transfer function can be only generated if the feedforward turns into positive feedback, likely to happen for the post #1 circuit.

    How do you measure the transfer function? It might be that a certain amount of load capacitance is required for stability, just fair for an OTA output stage. Check the literature in this regard.


    1 members found this post helpful.

    •   AltAdvertisement

        
       

  4. #4
    Newbie level 5
    Points: 139, Level: 1

    Join Date
    Sep 2019
    Posts
    9
    Helped
    0 / 0
    Points
    139
    Level
    1

    Re: RHP poles always appear in my designed amplifier

    Very thanks for your reply and simulation, I have checked the connection and they seems correct. I think maybe as FvM said bellow there is something wrong with my feedforward path and turns in to positive feedback.

    I am wondering any structure can drive capacitive load from 1p to tens of nF with high slew rate?

    - - - Updated - - -

    Quote Originally Posted by FvM View Post
    A RHP pole in the open loop transfer function can be only generated if the feedforward turns into positive feedback, likely to happen for the post #1 circuit.

    How do you measure the transfer function? It might be that a certain amount of load capacitance is required for stability, just fair for an OTA output stage. Check the literature in this regard.
    Here is the test bench I use, I found it from Allen P.E., Holberg D.R.'s book
    Attachment 157810

    I have tried to change the load capacitance from 1p to 50n, and it always oscillate. I am not Familiar with feedforward, I just use the structure from journals, and they did not mention anything about RHP poles.

    I also found the RHP poles are highly related to the first differential stage (adjust its ro and gm would have big affect on the number of RHP poles)
    Should this kind of amplifier so hard to adjust? is there any book mentioned about the basic knowledge of it, and how to eliminate the RHP poles?



--[[ ]]--