• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Subscribers 30 subscribers
  • Views 638 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

Current drawn for Speaker VDD in Class D amp

Titusc
Intellectual 660 points

Hi there,

I am thinking of using a class-D stereo speaker amp to drive two 4ohm 2W (normal) speakers.  Just wondering the current need for the power supply for the speaker VDD.  Do we calculate for the max current or the average current when using the power supply for the speaker VDD?

Assuming that I have to drive the 4ohm 2W speakers, and my speaker VDD is at 5V, I have (5V*0.707)^2/4ohm * 0.85(efficiency of the class D amp)=2.65W delivered to each of the speaker.  Now this calculation is based on a pure sine wave.  What if the wave form is such that a maximum load is reached?  And how do I spec for the power supply?  how much current should the supply provide?

thanks,

TC

  • 0
    TI__Genius 9175 points

    Hi TC,

    Since all of the power must come from the supply, I believe the supply needs to be designed to deliver the maximum RMS power, plus a safety margin.

    See HERE and HERE which may aid in your design process.

    I am going to forward this to a colleague who may be able to help more.

    Regards,

    -Matt

     

  • 0 in reply to Matthew Beardsworth
    Intellectual 660 points

    Thanks for the forward Matt.   Those are pretty comprehensive guides.

  • 0 in reply to Matthew Beardsworth
    Intellectual 660 points

    Hi Matt,

    Do people provision the power supply to handle worst case peak power?

    What if you don't design that and someone decides to play something that would draw that much power?  What will give?  

    Thanks,

    TC

  • 0 in reply to Titusc
    TI__Genius 9175 points

    Hi TC,

    The short and sure fire answer is yes. Design for worst case and add a margin and you will have no issues. A simple linear regulator would work in your case since the power demand isn't that large.

    A more elongated answer:

    Your question becomes much more complex due to the dynamic nature of audio.

     If we were talking about perfect sine waves and your power supply became a limiting factor, you would see the wave form begin to flatten at the peaks and begin to "squash" causing distortion. This is known as “sag” since you are loading down or “sagging” the power supply.

    This comes in two flavors:

    1. Hitting the supply current draw limit

    2.  Having poor supply regulation so that when you do draw a lot of current, the rail voltage drops due to ohms law voltage drop. V = IR where R is the supplys soure impedance.

    However, say there was a large spike in output power due to the music being played. This could be well over the power supply’s capable power delivery due to the power storage in the supply capacitors! Then when the transient or loud passage subsides, if the output power to the speaker is below the power supply’s max capabilities, it has some extra juice to recharge the supply caps. However, if we drain the cap’s energy completely before the loud passage subsides, then we will relay back on the supply and hit its limit causing sag. This is the RC time constant aspect of power supply decoupling caps.

    If you look around on the web you will find lots on this subject.

    There is much to learn on the subject epically considering all the power supply verities: regulated vs unregulated, linear vs switch mode.

    Regards,

    -Matt

  • 0 in reply to Matthew Beardsworth
    TI__Genius 9175 points

    Hi TC,

    I found more information on the subject HERE.

     

    -Matt

  • 0 in reply to Matthew Beardsworth
    Intellectual 660 points

    Thanks for the feedback and the support documents.  I came across a lot of different info on the web so didn't quite know where to go.

    I am clear now that the caps on the supply rail should be able to withstand the transient spikes in the audio band.

    Now, say if I have a power budget for a 5V 4ohm THD+D <=10%, based on 80% efficiency, is 2.5W.  This translates to Irms=0.79A and Ipeak=1.12A per channel from the 5V rail.  

    I think it is safe to provide 1.25A to each speaker amp channel.

    Thanks again for your input!

    TC