• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 29 subscribers
  • Views 1371 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.

TPA3255: LC filter design for PBTL pre-filter vs post-filter

Evan Cornell
Expert 1380 points
Part Number: TPA3255

Does the LC filter design change at all between PBTL pre-filter and post-filter configurations, aside from the pre-filter inductors needing to handle twice the current of each of the post-filter inductors?

Mostly, I'm interested in whether the cutoff frequency or the overall filter response changes between the two configurations, assuming that no saturation current limits are being hit in the inductors. 

  • 0
    TI__Guru 61360 points

    Evan,

    You are exactly right! Nothing changes except for the current handling of the inductor. Everything else is the same.

    Regards,

    -Adam

    If your question is answered, please click "This resolved my issue".

  • 0 in reply to Adam Sidelsky
    Expert 1380 points

    Adam,

    I have just completed some experiments that show that the frequency response is NOT the same for pre-filter vs. post-filter PBTL arrangement.

    The attached image shows frequency sweep (relative to 10W @ 1kHz, 2ohm load, PVDD=24V) of both pre-filter and post-filter arrangement. 

    Green trace is pre-filter PBTL with 2x Wurth 7443630700 and yellow trace is post-filter PBTL, 4x Wurth 7443630700. I also re-ran a frequency sweep with paralleling 7443630700 in the pre-filter configuration, to emulate 3.5uH inductors. This matched the response of the post-filter 4x 7uH setup. 

    This tells me that the inductance of the LC filter is halved when in PBTL, post-filter connection mode. 

    Thanks,
    Evan

  • 0 in reply to Evan Cornell
    TI__Guru 61360 points
    Evan,

    As our usual application is Audio, we do our THD testing (across frequency and power) up to 20k and also use brickwall filters that match. Is this acceptable for you or do you have further concern?

    Regards,

    -Adam
  • 0 in reply to Adam Sidelsky
    Expert 1380 points
    Adam,

    Now that I know that the frequency response isn't the same between pre-filter and post-filter, I can adjust my inductances to suit.

    But, for the record, the LC-filter response of 4-inductor, post-filter PBTL connection does effectively put the inductances of each BTL in parallel, thus halving the inductance value. So please keep this in mind when calculating PBTL LC filter component values.

    Thanks,
    Evan