• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 49 subscribers
  • Views 396 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

OPA189: Switching Artifacts not replicable in TINA Simulation

Kevin Mondoloni
Prodigy 10 points
Part Number: OPA189
Other Parts Discussed in Thread: OPA182, OPA192,

I am using the opa189dr as an output buffer for a DAC on a low noise board:

I have noticed switching artifacts at 280kHz when measuring the analog output: 

I could not find this frequency in the documentation of the opamp or when I simulated it using the official TINA Simulator with the following basic circuit: 

Is there a way to see these spikes in a simulation so that I can tune e.g. a RC filter in the feedback of the opamp properly?

  • +1
    TI__Expert 4460 points

    Hi Kevin,

    The pulses you are seeing are mentioned in the datasheet in Section 8.3.3 Input Bias Current Clock Feedthrough, pg. 22. Chopper amplifiers experience periodic switching transients of their input bias currents due to their architecture.

    These transient pulses can be exaggerated if the equivalent impedances at the inverting and non-inverting input terminals do not match. Since the DAC you are using has an unbuffered ladder architecture, the equivalent impedance at the non-inverting input can change and will be difficult to match on the inverting input.

    The simplified TINA behavioral models do not encompass the complex transient behavior of the chopper amplifier, but model other parameters such as the bandwidth and slew-rate limitations. The effects of capacitive loads can also be easily simulated.

    For buffering a DAC with high precision, I would recommend using the chopper amplifier OPA182 or the linear amplifier OPA192, as these are highly accurate and are more optimized for the application. The OPA182 has smaller switching transients while the OPA192 has none.

    Let me know if you have any questions!

    Thanks,
    Lucas

  • 0 in reply to Lucas Burnette
    Prodigy 10 points

    Hi Lucas, thank you for your response.

    The frequency of the switching artifact is fixed in my case. Is there a way to predict at which frequency it would be if I were to replace the opamp with the ones you mentioned? At first I thought I could guess it from the voltage noise spectral density but it doesn't seem to be the case for opa189.

  • +1 in reply to Kevin Mondoloni
    Guru 140200 points

    Hi Kevin,

    I see a period of 7µs? This corresponds to a fundamental chopping frequency of 140kHz which pretty well matches with the small peak in figure 7-18 of datasheet.

    I think the OPA189 indeed sees matched impedances on the two inputs, if I correctly understand the datasheet of your ADC? In this case the remaining switching noise can only be furtherly reduced by the help of a low pass filter which has to be added to the output of OPA189.

    Since the ADC shows a zero error of up to 21bit which corresponds to an offset error voltage of 100µV, I think a non-chopping OPAmp may also do in your application? By this you would get rid of the switching noise.

    Kai

  • 0 in reply to kai klaas69
    Prodigy 10 points

    Hi Kai,

    True, 140kHz would make more sense looking at fig 7-18 but how come that the first harmonic (280kHz) is bigger than the fundamental (140kHz)?

    And I guess that the linear OPA192 Lucas mentioned, would be suffiecient with an offset drift of ±0.2 μV/°C for this DAC.

  • +1 in reply to Kevin Mondoloni
    TI__Guru 63860 points

    Kevin,

    There are two clock cycles for each auto-correction - this explains 140kHz vs 280kHz frequencies difference.  Having said that OPA192 is the way to go here.

  • +1 in reply to Kevin Mondoloni
    TI__Expert 4460 points

    Hi Kevin,

    I corrected a statement in my original reply:

    The OPA182 has smaller switching transients while the OPA192 has none.

    Thanks,
    Lucas