• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 30 subscribers
  • Views 432 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.

Original question:

Clock driven Class-D audio amplifier

SLAU508 Analog PWM Error Amplifier OPA2365 TLV3502

Jonathan Moran1
Prodigy 45 points

Hi All,

I'm interested in using the reference design in SLAU508 for a class D amplifier, as mentioned in section 2 of the application note.

I'm trying to understand the error amplifier and the feedback from the comparator output. From my understanding this feedback will modulate the signal entering U1A with the aim of decoupling the final signal from U2A's rail noise/fluctuations. I can see how this would be beneficial if the signal was then filtered back to an analog signal using the output voltage of U2A however if this is fed into a Schmitt trigger / isolated gate drive the output should already be decoupled from U2A's rail and this feedback path could actually couple more noise onto the final output signal.

This is my current understanding, have I missed something?

Best Regards,

Jonathan

  • 0
    TI__Genius 15555 points

    Hello Jonathan,

    The error amplifier doesn't only decouple the average output voltage from the comparator's supply noise, but rather allows the average output to be a direct function of your input signal. This is best highlighted when comparing output voltage equations 2 & 3 as shown below:

    Notice that without the error amplifier, the average output voltage of the circuit is directly affected by the chosen amplitude of your triangle wave and PWM circuit. Since the power supplies fed into a comparator determines the amplitude of its output, the supplies used here will affect your average output voltage. While removing the effects of your comparator supplies noise is one of the benefits of the error amp, removing the VPWM element from the output voltage equation is the main advantage. 

    Best,

    Hasan Babiker

  • 0 in reply to Hasan Babiker31
    Prodigy 45 points

    Hi Hasan,

    Thanks you for your response, this does make sense as the main advantage.

    However, consider the case where I do have ripple/noise on U2A power rail, I understand that with the error amplifier feedback the filtered Vpwm will be decoupled from the power supply noise. If this PWM signal was instead used to drive a gate driver schmitt trigger, wouldn't the decoupling component taken away using the feedback loop affect the final driving PWM signal?

    Many Thanks,

    Jonathan

  • 0 in reply to Jonathan Moran1
    TI__Genius 15555 points

    Hello Jonathan,

    I'm not sure that I follow your question...

    The output elements we are discussing here are the average output voltage (or the analog signal) & the PWM output (or the digital signal). The purpose of the error amplifier is to reduce the effects of the comparator power supply to the average output voltage of your PWM output. The noise of your power supply reflected in your PWM signal output will still be determined by the PSRR of your comparator.

    If you were to follow up this design with a Schmidt trigger that had a cleaner supply or better PSRR specs than U2A, then you would reduce the effects of noise from U2A's supplies on your PWM output. 

    Best,

    Hasan Babiker

  • 0 in reply to Hasan Babiker31
    Prodigy 45 points

    Hi Hasan,

    Thanks again for your patience with this question,

    Yes ok I understand how the error amplifier reduces the effects of the comparator PSU on the average voltage of the comparator output signal when this PWM signal is directly filtered.

    My question is concerning the effects of this compensation when the output PWM signal is fed into schmitt trigger and isolated gate drive into push-pull running from a separate (HV) supply. In this case the PWM voltage signal from the comparator rail voltages is not used/filtered, only the PWM timing thus the average voltage does not need correction from U2A's rail noise.

    The comparator PWM output's high voltage and low voltage fluctuations from the comparator rails would not affect the final signal integrity yet the feedback compensation for the rail noise would still be included in the PWM signal.

    It seems like it would be desirable for the feedback to come from the output of the final push-pull drive instead however I am unsure how to modify the circuit for this

    Does that make sense?

    Best Regards,

    Jonathan

  • 0 in reply to Jonathan Moran1
    TI__Genius 15555 points

    Hello Jonathan,

    Yes, I think I understand your concern here. However it is important to note that the comparator supplies isn't the only element that is being removed from the average voltage equation with the use of a comparator feedback here. Without the feedback the average voltage is also being varied by the amplitude of your triangle wave & it's gain isn't as easily changed (when using feedback we can easily just adjust the gain using R3 & R4). 

    So if the supplies of the final PWM voltage is known, why not just adjust the gain of your average voltage using R3 & R4 to adjust?

    Best,

    Hasan Babiker