• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 31 subscribers
  • Views 753 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.

TPA3131D2: Bulk capacitor value selection

NIr Alon
Intellectual 490 points
Part Number: TPA3131D2
Other Parts Discussed in Thread: TPA3132D2

Hello all,

Regarding the value of the bulk capacitors required for PVCC:

In the product's PDF it is mentioned that (page 22)

"Select the bulk capacitors at the PVCC inputs for proper voltage margin and adequate capacitance to support the power requirements. In practice, with a well-designed power supply, two 100-μF, 50-V capacitors should be sufficient"

There is two versions of output power - 2X4W and 2X25W (TPA3131D2 or TPA3132D2)

The bulk capacitor value recommendation doesn't say for which power it refers to...

If I use only 2X4W, and components size is an issue in my design - Is there any way to calculate the actual required capacitance?

Thanks 

Nir 

  • 0
    TI__Mastermind 24835 points
    Hi Nir,
    For the bulk capacitors calcualtion, please refer to this document: www.ti.com/.../slyt199.pdf
    Best regards,
    Shawn Zheng
  • 0 in reply to Shawn Zheng
    Intellectual 490 points

    Hi Shawn,
    Thanks for your reply.
    I've looked into the application note - I need you to clarify equation 3 calculation:

    Equation 2 - Ipeak = sqrt of Ppeak/RL.

    In my case, 4W into 4ohm > Ipeak = 1A

    At the second equation I need your help to understand in which units to calculate the C value.

    Equation 3 - C>= (Ipeak * TD * Dmax) / Vripple

    Ipeak = 1A, Vripple = 100mV, fosc = 500KHz
    Vripple is in mV???
    TD is in %???
    TD is in KHz???

    I've tried this calculation, I assume it's OK - please verify:

    C>= [1 * (1/500*10^3) * 0.8] / 100*10^-3

    C = ~ 16uF per channel

    Is this correct?

    Thanks!

    Nir

  • 0 in reply to NIr Alon
    Guru 140200 points
    Hi Nir,

    why not following the recommendations given in figure 29 of datasheet? You can clearly see that the power supply decoupling scheme is recommended for both, the TPA3131D2 AND the TPA3132D2.

    Kai
  • 0 in reply to kai klaas69
    Intellectual 490 points

    Hi Kai,
    220uF Caps are huge, and they probably large in order to support 25W operation (TPA3132D2)
    In my design I use only 4W (even less, my power supply is only 5Vmax) so I see no reason to use a capacitor as large as being used in the application note.
    Nir

  • 0 in reply to NIr Alon
    TI__Mastermind 24835 points

    Hi Nir.

    Let's take Dmax=1 for magain and we suppose 50mV ripple is accepatable. Then the capacitor should be 50uF for 400kHz PWM frequency. We usually reoommend users to user to use at less 100uF bulk capacitor.

    Best regards,

    Shawn Zheng

  • 0 in reply to NIr Alon
    Guru 140200 points
    Hi Nir,

    It's not only the capacitance what counts but also the ESR of aluminium electrolytic. Here is the ESR of some FC Panasonic electrolytics:

    220µ/50V, ESR = 0.09R
    100µ/50V, ESR = 0.162R
    68µ/50V, ESR = 0.234R
    68µ/16V, ESR = 0.8R

    It's true that the ESR of 68µ/16V is much smaller than the 220µ/50V. But its ESR is about ten times higher! This means that you inject about ten times more 400kHz switching noise into your 5V supply voltage.

    So, it's not wise to use a much lower capacitance and a much lower voltage rating.

    Kai
  • 0 in reply to kai klaas69
    TI__Mastermind 24835 points
    Hi Nir,
    Kai is right. Generally ESR should be less than Vripple/Ipeak, here Vripple means the desired ripple voltage and Ipeak mean the peak current. And the capacitor ESR should be 30~50% lower than this, condisering temperature, ESL and aging.
    Best regards,
    Shawn Zheng
  • 0 in reply to Shawn Zheng
    Intellectual 490 points

    Hi Shawn and kai,

    Thanks for all your comments.

    What you are saying is true, but I didn't mentioned that I'm not going to use electrolytic capacitors at all, so ESR won't be an issue.

    I plan to use a 1210 size 100uF/6.3V cermaic capacitor (the actual capacitance @ 5V will be ~40uF) 

    Shawn - your calculation is for 4W / 4ohm, In my case Pout will actually be less, my power supply is 5V max and to get 4W I will need at least 7.5V - so I believe that 40uF will be sufficient.

    If I had to use 220uF caps like recommended in the PDF, I probably couldn't use ceramic capacitor due to their large size.

    Anyway, I will this on the EVB before jumping into conclusions.

    Thank again

    Nir 

     

     

  • 0 in reply to NIr Alon
    TI__Mastermind 24835 points
    Hi Nir,
    Thanks for your update. It's a good idea to do some experiments on the EVM at first. Please feel free to reply in this post if you need any help.
    Best regards,
    Shawn Zheng