Post Go back to editing

Output Common mode voltage is not centered at 1v

Thread Summary

The user is unable to achieve the desired output common mode voltage (Vocm) of 1V with the ADA4937 when converting a 2.25Vpp single-ended input to a 2Vpp differential output to drive the AD9255 ADC. The datasheet indicates that Vocm cannot be below 1.2V. The user should use the Diff Amp Calc tool to determine appropriate resistor values and ensure the circuit topology is correctly configured in LTspice. The user also uploaded an LTspice schematic file for further review.
AI Generated Content
Category: Software
Product Number: ADA4937-1
Software Version: LTspeice (x64) 24.1.10

ADA4937 is used to Convert my SE input (2.25vpp) to differential output( 2vpp)  with vocm of 1v to drive the ADC AD9255. But i am unable to get the output common mode voltage of 1v. but instead i am getting 2.16v vocm. Also i have included a (62.5Mhz) anti- alias filter between Amplifier output and ADC.    Please add steps someone would take to reproduce your experience.. What mistake did i made? 

Thread Notes

  •  Please attach your schematic file (.asc) so we can take a closer look?

    -Anne

  • Project_Ref.asci have uploaded (.asc) file.. Although i have make a path for DC to flow at the output side, still i am getting output common mode voltage 2.16v.. 
     

  •  Take a look at the datasheet, as well as https://www.analog.com/en/resources/interactive-design-tools/adi-diffampcalc.html

    The datasheet indicates that Vocm can't be below 1.2V:

    Also - take a look at Diff Amp Calc to get an idea of what the voltage limitations are for the circuit you are trying to implement.  Here's a snapshot of where I'm trying to replicate your circuit in LTspice:

    Note that I've selected a single-ended topology, terminated at the input, and AC Coupled 2 at the input.

    Take a look at Diff Amp Calc, figure out a circuit topology that you're happy with, and then modify your LTspice schematic to match.  I hope this is helpful.

    -Anne

  • Hi Anne,

    Thanks for the detailed response and for sharing the Diff Amp Calc example. I’ll review the tool and modify my LTspice schematic accordingly to match the single-ended, AC-coupled topology you mentioned.

    I noticed the resistor values are shown clearly in your schematic, but could you please clarify the AC coupling capacitor values used at the input? I’d like to ensure my circuit maintains the correct low-frequency response and matches the AD4937’s voltage limits.

    Thanks again for your guidance!
    – Gowtham

  • Hello Anna 

    I have simulated the version you have provided 
    here are the results.... 

    I have observed with 5v supply i am getting mid point as VOCM (2.5 or 2.6v ).. 

    the DC operating points: 
    V(o+): 3.45273 voltage
    V(o-): 1.10845 voltage
    V(vinp): 0.000140437 voltage
    V(vn): 0.00144786 voltage
    V(vocm): 1.2 voltage
    V(vp): 0.00144786 voltage

    i am attaching the (.asc) file.. suggested Project.asc

    I didn't understand why the output common mode is always getting mid point of the supply voltage instead of suggested 1.2v Dc. Could please verify or may be suggest an solution..  

    Thanks for you time and efforts.  

  •  Thanks so much for the updated schematic...

    I see the issue you're having, and I referred to the example circuit for the model to verify that the model *should* be working fine.  Here's what I see with the example circuit:

    The interesting thing I noticed with this example is that it doesn't make use of the FB pins at all.  So, I modified your schematic to connect the feedback network the same way:

    Common mode is now looking better but it looks like the output is clipping near 1V, so I increase the common mode to 2V:

    This seems to be behaving as I expect.

    Now - taking a step back and looking at what those FB pins are even for...  From the datasheet:

    So - if you were to use those FB pins in your simulation, here's how you would connect them:

    These FB pins are SUPER handy for layout/routing considerations on your PCB, but aren't all that meaningful in a simulation schematic - so I'll leave it to you as a personal preference if you want to connect OUT to IN of FB to IN in your LTspice schematic.

    I hope this helps.

    -Anne

  • hello Anne 

    I just verified and re-wired my schematic accordingly. Now the feedback connections are working as excepted. 

    Thanks for you help. In my screenshot i had noticed that the differential swing is around 0v.. for 1.2 vocm. 

    Thanks a lot  again for you help and guidance and kind responds