+ Post New Thread
Results 1 to 7 of 7
  1. #1
    Member level 1
    Points: 214, Level: 2

    Join Date
    Feb 2019
    Posts
    34
    Helped
    1 / 1
    Points
    214
    Level
    2

    Simple op-amp current sink question

    Hi,
    I'm working on a battery charger using an LTC4020 controller and one of the inputs allows scaling of the charging current.

    The RNG/SS pin is described as follows:
    Click image for larger version. 

Name:	SS.jpg 
Views:	8 
Size:	263.9 KB 
ID:	153180

    Later in the datasheet, they recommend this method for controlling the voltage on the pin from an active servo capable of only sinking current:
    Click image for larger version. 

Name:	SS_app.jpg 
Views:	7 
Size:	120.2 KB 
ID:	153181

    I have done what's shows in the schematic but it does not work correctly. Here is my PCB schematic:
    Click image for larger version. 

Name:	sch.jpg 
Views:	16 
Size:	133.4 KB 
ID:	153182

    The RNG signal is from a DAC and outputs the correct voltage (0-1V) When it is set to 1.0V, the inverting input of the op-amp (Pin 4) has 1.0V, but the non-inverting input (and the RNG/SS pin) have about 0.2V

    Is there is a reason this circuit is not working for me? When I remove R34, the charger works correctly at full current, but I need to throttle the current down.

    •   AltAdvertisement

        
       

  2. #2
    Advanced Member level 5
    Points: 38,969, Level: 48

    Join Date
    Mar 2008
    Location
    USA
    Posts
    6,300
    Helped
    1834 / 1834
    Points
    38,969
    Level
    48

    Re: Simple op-amp current sink question

    SS pins usually have a pretty low current source value
    and maybe the FET leaks so much that it can't be driven
    "off" well enough by the op amp VOL, to let the pin rise
    up.


    Have you looked at what's on the op amp output, vs the
    FET's advertised VT range and the current at that VT, to
    see what the FET's being told to do?

    Also some possibility that the local loop is not stable, what
    happens if you are oscillating on SS pin? Are you using an
    oscilloscope, or a DMM to get these voltages? Op amps
    are not really up to driving nF-range loads generally, you
    may need to add a trick or two to get stable behavior.



    •   AltAdvertisement

        
       

  3. #3
    Advanced Member level 1
    Points: 2,387, Level: 11

    Join Date
    Aug 2016
    Posts
    414
    Helped
    65 / 65
    Points
    2,387
    Level
    11

    Re: Simple op-amp current sink question

    Hi,

    I just noticed that you have two different GND on the schematic you posted. Any reason for that?
    -------------
    --Akanimo.



    •   AltAdvertisement

        
       

  4. #4
    Super Moderator
    Points: 74,683, Level: 66
    Achievements:
    7 years registered
    Awards:
    Most Frequent Poster 3rd Helpful Member

    Join Date
    Apr 2014
    Posts
    15,145
    Helped
    3448 / 3448
    Points
    74,683
    Level
    66

    Re: Simple op-amp current sink question

    Hi,

    I think DMN2005 is not a suitable Mosfet.
    It is capable to drive 100A, but the Current in question is 50uA only. (A deviation of 2 million).
    Better use a small signal (Mos)fet.
    An R_ds_ON of 20Ohms or less should be suitable.

    I'd also omit the capacitor, it may harm stability. (If you really need it you should consider to use a suitable RC combination)

    This is where I see the problem. I don't think it is a DC problem, but an oscillating problem.
    Use a scope to check this.

    Klaus
    Please don´t contact me via PM, because there is no time to respond to them. No friend requests. Thank you.


    1 members found this post helpful.

  5. #5
    Advanced Member level 3
    Points: 5,169, Level: 17

    Join Date
    Feb 2014
    Posts
    714
    Helped
    243 / 243
    Points
    5,169
    Level
    17

    Re: Simple op-amp current sink question

    Yeah Omit the cap on the fet output. Choose a small signal fet and choose a lower bandwidth opamp or compensate the opamp for lower bandwidth (2mhz or less is what I’d think, and perhaps way less because you don’t need speed here).


    1 members found this post helpful.

  6. #6
    Member level 1
    Points: 214, Level: 2

    Join Date
    Feb 2019
    Posts
    34
    Helped
    1 / 1
    Points
    214
    Level
    2

    Re: Simple op-amp current sink question

    Thanks all.

    The separate SGND and PGND are recommended in the LTC4020 datasheet (it has pins for both and they are net-tied near the chip). The SS pin is in reference to SGND.

    I'll remove the capacitor and take new measurements with a scope and look into different components.

    I just wanted to make sure the generic circuit is correct before diving into it very far. In terms of voltage followers, I've only found examples where the Vin voltage connects to the non-inverting (+) input and it is a current sourcing circuit. The datasheet also doesn't specify using a n-mos or p-mos FET but n-mos seemed logical. Thanks.

    - - - Updated - - -

    I have replaced the FET and op-amp with new parts and removed the capacitor and it now works as intended. I can set the SS pin from 0-1V now so I think it was the capacitor causing stability issues.

    BTW, Klaus, the FET I'm using is DMN2005K which has ID max of 300mA. Sorry, I should have corrected earlier.
    https://www.diodes.com/assets/Datasheets/ds30734.pdf

    Is there a better choice small signal FET in the same package (SOT-23)?



    •   AltAdvertisement

        
       

  7. #7
    Super Moderator
    Points: 74,683, Level: 66
    Achievements:
    7 years registered
    Awards:
    Most Frequent Poster 3rd Helpful Member

    Join Date
    Apr 2014
    Posts
    15,145
    Helped
    3448 / 3448
    Points
    74,683
    Level
    66

    Re: Simple op-amp current sink question

    Hi,

    BTW, Klaus, the FET I'm using is DMN2005K which has ID max of 300mA. Sorry, I should have corrected earlier.
    https://www.diodes.com/assets/Datasheets/ds30734.pdf

    Is there a better choice small signal FET in the same package (SOT-23)?
    What a big difference with a suffix can make...

    Choice: just use a simple, cheap Mosfet. It should work.

    Klaus
    Please don´t contact me via PM, because there is no time to respond to them. No friend requests. Thank you.



--[[ ]]--