• State Resolved
  • Locked Locked
  • Replies 28 replies
  • Subscribers 31 subscribers
  • Views 5335 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.

Designs based PCM5122 and TPA6133

Andre Strul
Expert 1330 points
Other Parts Discussed in Thread: PCM5122, TPA6133A2, TPA6133A2EVM

hello to all,

we have some doubts concerning our design based on the above mentioned chips...

Thank you in advance for your feedback.

  1. PCM5122 DAC:-                                                                On Audio output  R C circuit(22pF) , Capacitor should be Ceramic capacitor or film capacitor?
  2. TPA6133 head phone amplifier:                                Input DC Blocking Capacitor on audio path  (.47uf) should be Ceramic capacitor or film capacitor?

regards

  • 0
    TI__Guru 57800 points

    Hi Andre,

    You can use Ceramic capacitors on the input of the TPA6133A2. I have asked my colleague to address the question for the PCM5122.

    Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • 0
    TI__Mastermind 27850 points
    Hi Andre,

    We use a ceramic capacitor on the EVM for the PCM5122.

    Justin
  • 0 in reply to Justin Bohr
    Expert 1330 points
    hello and thank you to both of you.
    i guess the correct question would be....
    what should we use in order to achieve the highest sound quality possible?
    film or ceramic ?

    thank you
  • 0 in reply to Diego Melendez
    Expert 1330 points
    Hello Diego,

    can you please give us the value for the following capacitors on the TPA6133A2EVM board from TI.
    (C10,C11,C12,C13,C15,C16,C17,C18,LOAD_L,LOAD_R)


    thank you
  • 0 in reply to Andre Strul
    TI__Guru 57800 points

    Hello Andre,

    Those components are used to add filtering options for the evaluation of the device:

    C10, C11, C12, C13, R3, R5 are used as an optional low pass filter used to measure the output of the amplifier with audio analyzer as 9.2.2.4 Optional Test Setup section indicates.

    C15, C16, C17, C18, R9, R10, R11, R12 are used as an optional High pass filter that can be used when the input of the amplifier is connected to the output of a codec, to reduce any out of band noise. 

    LOAD_L, LOAD_R can be used to populate a load to simulate the speaker that the amplifier will drive. 

    The values of these components may be adjusted depending on the desired cut off frequency for each filter.

    Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • 0 in reply to Andre Strul
    Expert 1330 points
    hello to both.
    i wanted to know if you can please let me know your views on the ceramic vs thin film.
    what should we use in order to achieve the highest sound quality possible.

    thanks again .

    regards,
  • 0 in reply to Andre Strul
    TI__Mastermind 27850 points
    Hi Andre,

    The main difference for film vs ceramic capacitors is the temperature response. Here is a page where Panasonic has done a comparison here:

    industrial.panasonic.com/.../02

    Justin
  • 0 in reply to Diego Melendez
    Expert 1330 points
    Hello Diego,

    can we develop a small PCB with the components above only for testing purposes ? (connected to tpa3116 board).

    Our PCB space is limited...
    how would we connect?

    thank you

    Andre
  • 0 in reply to Diego Melendez
    Expert 1330 points

    hello Diego

    kindly answer our last question when you can.

    thank you

  • 0 in reply to Andre Strul
    TI__Guru 57800 points

    Hi Andre,

    I would recommend to follow the layout guidelines from the datasheet of these devices. Can you please specify the board connections available for testing or provide a block diagram so we can have a better idea to direct you with the correct connection?

     Regards,

      -Diego Meléndez López
       Audio Applications Engineer

  • 0 in reply to Diego Melendez
    Expert 1330 points

    thank you Diego!

  • 0 in reply to Diego Melendez
    Expert 1330 points
    Hello Diego.
    the block diagram is ready...
    how can I attach a pdf here?
    thank you
    Andre
  • 0 in reply to Andre Strul
    Expert 1330 points
    if you can send me an email i can email it to you
    andre at allo dot com
  • 0 in reply to Andre Strul
    TI__Guru 57800 points

    Hi Andre, 

    You can find information on how to post a file ion this page:

    https://e2e.ti.com/group/helpcentral/w/e2e/148.4-5-attaching-a-file

    If you are still having problems, please send me an email (diego.melendez@ti.com).

    Best Regards,

      -Diego Meléndez López
       Audio Applications Engineer

     

  • 0 in reply to Diego Melendez
    Prodigy 230 points

    On the capacitor question, these 2.2nF (NOT 22pF) are effectively filter caps and as such have significant signal voltage across them. 

    Had they been 22pF then C0G/NP0 ceramic would have been the way to go, but that technology does not extend to nF values, and the ceramics that do work up there (X7R) typically have enough voltage non linearity to hurt the harmonic distortion numbers at high output levels, go for film if your target BOM cost can stand it.

    Were these coupling as opposed to filtering caps then the key is to make them large (so that the signal voltage appearing across them is negligible), and then anything that is sufficiently low leakage goes (NP Elco is common).

    It is a little old now, and Electrolytic caps in particular have gotten much better, but Cyril Bateman write the classic series on capacitors and distortion in audio for E&WW some years ago, you can find it here.

    73 Dan.

     

     

  • 0 in reply to Justin Bohr
    Expert 1330 points
    Justin,

    We are thinking on using a LPF filter (LC) on the power supply of the DAC (to remove high frequency noise) with a cutoff frequency of 100Khz-150Khz

    Instead of that we can also use the DSP function of the PCM5122 to make a LPF (with same cutoff frequency)

    What do you recommend ? If both filters are applied do you see any advantage ?

    thank you
  • 0 in reply to Andre Strul
    Expert 1330 points
    Hello Justin.
    Sorry to bother again, i was wondering if you can please give us a reply on the question above.

    thank you
  • 0 in reply to Andre Strul
    TI__Mastermind 27850 points
    Hi Andre,

    Having the physical LPF filter will help eliminate any noise from the board/output stage from the device. So it will be best to use an external LPF is the noise you are trying to eliminate is not in the audio itself.

    Justin
  • 0 in reply to Justin Bohr
    Expert 1330 points
    Thank you Justin
  • 0 in reply to Diego Melendez
    Expert 1330 points
    Hello Diego,

    please support us with the following:

    Our requirement

    1. To operate the PCM5122 DAC with Clock Auto mode set in slave mode.
    2. We will connect an external clock to SCK input (pin 20).
    3. We are using for only Common audio sampling frequencies.
    4. We want the DAC to use this external clock(SCK) as the reference for all other clock generation with Clock auto mode set.

    Clarification required:

    1. If clock auto mode is set and if both SCK and BCK are present then will the DAC choose BCK or SCK as the reference.
    2. If it chooses BCK then how to make the DAC choose the SCK as the reference clock with Clock auto mode set.

    thank you!
  • 0 in reply to Andre Strul
    TI__Mastermind 27850 points
    Hi Andre,

    1. If both SCK and BCK is supplied, the PCM5122 will automatically use the SCK. If SCK is at GND while good clocks are on BCK and LRCK, then the PCM5122 will automatically configure to use BCK and LRCK.

    2. You will only have to choose what clocks go to the DAC if you are not using auto clock configure mode.

    Justin
  • 0 in reply to Justin Bohr
    Expert 1330 points
    Dear Justin,

    thank you for the previous reply.
  • 0 in reply to Justin Bohr
    Expert 1330 points

    Hello again.

    see follow up question.

    We are using PCM5122 DAC. The DAC is connect to our SOC via I2S.

     

    1. When we play audio file 16bit/24bit/32bit the SOC always sends 32 bit data to the DAC.

    In case of 16 and 24 bit it is padded(with zeros) to make it 32 bit.

    does the output sound quality degrade because of this padding?

     

    1. Will the sound quality be better when we send, 16bits or 24bits only when playing from 16bit or 24bit file without padding.  (example  16bit file 16 bits of data transferred with 16 BCLK clocks or 24bit file 24 bits of data transferred with 24 BCLK clocks).

     

    thank you for your support!

    Andre

  • 0 in reply to Andre Strul
    TI__Mastermind 27850 points
    Hi Andre,

    There a re a few limits that are in play. First you have the quality of the audio recording (resolution and SNR of equipment used to record)that will limit the best theoretical SNR limit. Then you have the resolution of the I2S connection you are using to transfer this audio data.

    If the audio was recorded with an ADC at 16 bit resolution, or the SNR of the ADC was -96 dB or worse, then there will be no difference playing the audio content at 16 bit, 24 bit or 32 bit. However, if the ADC used for the recording used 24 bit resolution and had an SNR of something above -96 dB then you would be loosing out on audio accuracy if it was played using 16 bit I2S.

    The SoC most likely send 32 bits to the device at every resolution because the clocking requires it. If the BCK to LRCK ratio is 64, then no matter the resolution, there needs to be 32 bits per channel. If the I2S is operating at 16 bit resolution, this will be 16 bits of audio content and 16 bits of padding (0s).

    In summary, if your audio was recorded at high enough quality, then playing at a lower resolution will effect the audio quality. If it was recorded at 16 bit resolution, playing at 16 bit or higher will have the same quality.

    Justin
  • 0 in reply to Justin Bohr
    Expert 1330 points

    Dear Justin and team,

    We need to modify the capacitor to higher value for our piano board (Reference circuit below).

    PCM5122 IC analog path Capacitor:

    1. how much higher value we can go from current value10uF ?

    2. will there be any compatibility issue arises, if we use higher value Capacitor for our circuit design ?

    thank you

  • 0 in reply to Andre Strul
    Expert 1330 points
    hello,
    please let me know if you guys had a chance to look into this.
    thanks and have a great Monday.

    Andre
  • 0 in reply to Andre Strul
    TI__Mastermind 27850 points
    Hi Andre,

    Using a larger bulk capacitor for AVDD should not cause any issues, but might create a slower startup.

    Justin