The unusual distortion of AD8031

Dear Sir/Madam,

We would like to use AD8031ANZ for  a unity gain follower.
Pls refer to the schematic. below,

There are description on the datasheet.
/P.14:
Used as a unity gain follower, the output of the AD8031/
AD8032 exhibits more distortion in the peak output voltage
region around VCC − 0.7 V. This unusual distortion  is
caused by the input stage architecture....

/

However,we could not understand the  "unusual distortion".

What kind of distortion?.

Pls give advice to us.

Thank you.

Best Regards,
Kou

  • 0
    •  Analog Employees 
    on Sep 29, 2013 11:00 PM

    Hi Kou,

    Thank you for that very interesting question. The "unusual distortion" really is due to the rail-to-rail input (RRI) property of AD8031 which uses dual differential pair at the input stage. For non-RRI op amps, a single differential pair will be sufficient to cover input common mode range of up to the positive or negative supply, but not both. For the op amp to cover input range from positive to negative supply, two differential pair are used, just as Figure 43 of the AD8031 datasheet shows.

    However, this topology has its drawbacks too. First, that this topology introduces bias current reversal as shown in Figure 11 of the datasheet.

    As we can see, the bias current changes direction as we approach common mode of Vcc-1.1V. This will translate into a change in polarity in the output offset due to bias currents, which is a form of distortion. One method to reduce the effect of bias current reversal is to use lower value resistors. You mentioned that you want to implement a unity gain follower, is there a specific reason why you chose to put some resistors instead of the simple unity gain buffer configuration where the output is directly connected to the non-inverting input? And just a side note, the 10k and 1k resistors contribute 12.65nV/rtHz and 4nV/rtHz noise, respectively, which significantly raises your noise floor.

    The second drawback of the input topology is an introduction of crossover distortion at the input. This is shown on figure 12 of the datasheet.

    The input offset voltage changes its value somewhere near the Vcc-1.1V input common-mode region. And similar to the bias current reversal, this change in input offset voltage contributes to the overall non-linearity of the device.

    These factors are just some of the reason why other engineers don't prefer RR op amps. I am not sure what your expected input voltage range is, but if you can do away from going beyond Vcc-1.1V, which in your case is 2.2V, then you won't have to deal about the distortion I discussed. But if you can't, we'll try to find some other options.

    I hope this help.

    Best regards,

    Neil

  • 0
    •  Analog Employees 
    on Sep 30, 2013 5:22 PM

    You're very welcome! And by the way, the dual differential pair is just one architecture to have rail-to-rail inputs, just an FYI. Good luck!

    I hope this has been informative to you. I would appreciate if you could mark my response as "Correct" or "Helpful."

    Thank you.

    Neil

  • Hello Neil,


    Thank you for making the time for my question.

    I am going to learn RtoR opamp more.

    Regards,
    Kou

  • 0
    •  Analog Employees 
    on Aug 2, 2018 2:27 PM
    This question has been assumed as answered either offline via email or with a multi-part answer. This question has now been closed out. If you have an inquiry related to this topic please post a new question in the applicable product forum.

    Thank you,
    EZ Admin