• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 32 subscribers
  • Views 537 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.

TAS5756M: Some basic questions

Greg Valvo
Prodigy 30 points

Part Number: TAS5756M
Other Parts Discussed in Thread: TAS5805

So, starting a new board development and it looks like TAS5756M may be a good fit. I've been looking through the datasheet and have a couple really basic questions that I hope to get answered.

1. The device has many DSP features that allow extensive Audio Chain customization. This is done through "HybridFlows". Choosing the configuration parameters for a "HybridFlow" is so complex that it is done through TI's proprietary "PurePat Console" software and downloaded to an expensive TI-proprietary Eval Board. Correct so far?

That's all well and good for evaluating the chip, but how does it work on my production board? How is a "HybridFlow" saved and loaded into the device on my board? Non-volatile memory within the TAS5756M and loaded on power-up? Non-volatile memory on my board and loaded over I2C by a separate processors? The TAS5756M acts as a bus master to download the "HybridFlow" from an external I2C EEPROM on my board (without an external processor)?

How is the register configuration for a chosen "HybridFlow" extracted from "PurePat Console" and used?

2. From Section 8.3.1 of the datasheet (Power-on-Reset (POR) Function), I note the following paragraph:

"The TAS5756M device has a power-on reset function. This feature resets all of the registers to their default configuration as the device is powering up. When the low-voltage power supply used to power DVDD, AVDD, and CPVDD exceeds the POR threshold, the device holds sets all of the internal registers to their default values and holds them there until it receives valid MCLK, SCLK, and LRCK/FS toggling for a period of approximately 4 ms. After the toggling period has passed, the internal reset of the registers is removed and the registers can be programmed via the I2C Control Port."

That seems like a relatively straight forward statement, but I find it hard to believe. Is it saying that the device's registers can't be configured over I2C until a valid I2S input signal (MCLK, SCLK, LRCK) is present? What would be the rational for such a design decision? It makes much more sense to me to be able to configure the device on power up with or without an I2S signal and then go from there. I certainly hope someone will tell me that my interpretation of the above paragraph is incorrect.

Thanks all.

  • 0
    TI__Genius 9995 points

    Hello Calvo,

    Can you list your basic specifications for your design? I am trying to recommend TAS5825 (2*38W) or TAS5805(2*23W) for you as they are the up-to-data family that has been widely used around the world. 

    As to TAS5756, it is kind of old and the software is based on last gen PPC2 that will not be updated. TAS5825/TAS5805 is based on PPC3 and it is more powerful than TAS5756 in both DSP and design GUI. 

    As to your questions, my reply is as below:

    1. Yes. It is suggested to use TI EVM with PPC2. When you finish configuring TAS5756, you can dump all the settings with PPC2 and then your processor in your board can download these configurations into the TAS5756 on your board by I2C. Further, with TI EVM board, you can jump I2C directly to your board, then you can tuning through GUI directly in your board. (This function works better in PPC3 as we support optional I2C address ). Finally, you can use EEPROM to configure TAS5756. As to TAS5825, we have a guide for SPI: http://www.ti.com/lit/an/slaa847/slaa847.pdf . As to how to dump, you can refer to related materials by :http://www.ti.com/product/TAS5756M/technicaldocuments
    2. I2C configuration is needed to be under valid I2S. This decision is determined by the clock tree of the device, which you can find on the datasheet.

    Regards,

    Alix Wan.

  • 0
    Prodigy 30 points

    Hello. Thank you for the reply. If I may paraphrase it to confirm that I fully understand:

    1. The TAS5756 contains no internal non-volatile memory to save a custom configuration. So, the configuration must be downloaded to the device over I2C every time it is powered up.

    2. The I2C configuration download can only take place when a valid I2S input signal is present.

    Please confirm that the above is correct or provide any necessary correction.

    Thanks again.

     

  • 0 in reply to Alix Wan
    Prodigy 30 points

    Hello. Thank you for the reply. If I may paraphrase it to confirm that I fully understand:

    1. The TAS5756 contains no internal non-volatile memory to save a custom configuration. So, the configuration must be downloaded to the device over I2C every time it is powered up.

    2. The I2C configuration download can only take place when a valid I2S input signal is present.

    Please confirm that the above is correct or provide any necessary correction.

    Thanks again.

  • 0 in reply to Greg Valvo
    TI__Genius 9995 points

    Hello Calvo,

    TAS5756's configuration will be lost every time it powers off. Hence, it needs to be re-initialized once upon it powers from off to on.  As to:  The I2C configuration download can only take place when a valid I2S input signal is present, it is correct.

     

    Regards,

    Alix Wan.