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.

TPA3116D2: Power supply Pin connections

Part Number: TPA3116D2
Other Parts Discussed in Thread: TPA3128D2

Hello,

We are currently working on a design for an amplifier using TPA3116 to be powered of nominal 12V with an internal boost converter to generate 24V on demand (possibly using signal envelope tracking) if input signal demands require this. TI does have lovely reference designs for this.

However we already use and deploy suitable integrated step-up converter solutions, where one pair of converters in parallel would in principle scale sufficiently to deliver close to rated power. To us design reuse and economics of scale matter a good deal here, so we would like to employ this scheme in preference over a single higher power converter.

Obviously directly paralleling power supplies (even synchronized ones) is asking for trouble with current sharing.

However, it seems that each channels PVCC connections are pinned out separately (possibly even each half bridge - counting the pins). In this case we could use one of our existing solutions to power one channel, giving us the desired outcome.

Can TI confirm that the PVCC Pins are separated and may be connected to separate supplies?

Additionally, anecdotal evidence exists that extra filtering for the AVCC pin improves performance.

Are there any voltage difference limits between each channels PVCC pins and the AVCC Pin that must be respected, or are we reasonably free to set AVCC and PVCC Left and PVCC Right separately as long as maximum voltage limits are respected?

So could we (for arguments sake) supply AVCC with a well filtered 12V Supply and have the PVCC Pins at any value between (say) 9V and 26V depending on demand by the signal?

Or do we need to make sure (say back to back diodes) that the various supplies track closely?

Thank you for your consideration, T

  • Hi Thorsten,

    Thanks for asking questions in E2E. PVCC are connected inside the amplifier, so PVCC pins should be supllied with the same voltage. And AVCC is also connected to PVCC inside the amplifier, so AVCC should use the same voltage as PVCC. If you want the seperated AVCC power supply from PVCC, I think you try TPA3128D2, which has very low idle current.

     Best regards,

    Shawn Zheng 

  • Dear Shawn,

    Forgive me for saying so, but I have strong reasons to believe your answer is incorrect on the facts of the matter. 

    It is well known that AVCC and PVCC for the TPA3116 are not internally linked and high performance applications of the TPA3116 commonly place an extra RC filter to the AVCC Pin and with this filter the Voltage at AVCC is slightly lower than at PVCC (naturally so).

    Using a multimeter the L & R PVCC pins also do not appear linked.

    I am aware of the information in the datasheet regarding these pins, however it is not explicit in any way.

    Further, regarding TPA3128, to my application idle current is immaterial, but I require a Power Pad Up device.

    The question here is actually how to supply the maximum current at full power using a 12V to 24V Step-up switcher.

    To deliver one high current 24V Step-up is possible, however, as I stated we have a strong preference to deploying our existing solution, which can provide around 40VA continuous, multiple times to achieve the desired power rating. If necessary I can design in suitable current sharing circuitry etc., however I feel using one lower power converter per channel would deliver fundamentally the same result.

    This would not only cut development time drastically, the ability to use identical parts across multiple products will aid cost reduction both by actually being less expensive than the single high power solution proposed by TI an and by economy of scales, reduced SKU's etc.

    I would greatly appreciate if you could ask Mr. Matthew Beardsworth to have a look at my question. He has been previously very helpful in enquiries about TI's Class-D amplifier parts.

  • Hi Thorsten,
    I chekced with the designers about this, and they confirmed the internal connection.
    So, you want the seperate power supply for the each channel but the voltage for the two channels is the same, right? We never saw this kind of application before, and it looks fine. But we can't assure the reliability.
  • Hi Shawn,

    Ok, I will spend some time tomorrow trying to fry me some fish'n'chips (TPA3116 Chips), have a look-see if the internal connections show up with sufficient voltage and current...

    In operation the Voltages will be ALMOST the same, but usual tolerances apply (< 3% on the Vref of our switch IC and 1% for resistors).

    The reason is that we have fairly standardised on a single step-up switch IC which we use widely and also on the (custom made) chokes that go with it. But the reliably available current at 24V from 12V is limited at 1.5A, enough for 1/2 TPA3116, but not for both channels. So the logical attempt to power the channels separate.

    I can instead make interleaved converters with a master/slave system of current sharing, but TBH, I don't really see the point.
  • Following more testing, PVCC Pins are internally linked. AVCC Pin is separate.

    I have hence re-designed the system to take account of the current sharing between multiple Step-up switchers.