• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 1 answer
  • Subscribers 30 subscribers
  • Views 211 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.

TAC5112: Request for Input Configuration Guidance – TAC5112 Codec with XLR and TRS Inputs

Shemaiah Rangitaawa
Prodigy 20 points
Part Number: TAC5112
Other Parts Discussed in Thread: TAC5212, INA217, INA333

Tool/software:

Hello,

I’m currently working on interfacing both XLR and TRS inputs with the TAC5112 codec and would appreciate some clarification regarding the recommended input configuration and best practices for each input type. I have a few specific questions:

  1. Bias Voltage
    From my understanding, bias voltage should only applied externally to microphone-level signals while instrument-level pickups are internally biased by the codec. Is that correct?

  2. XLR Input
    The XLR input provides phantom power for condenser microphones, and the input impedance is set using a 4.22 kΩ resistor. Would the XLR input path be compatible with the codec?

  3. ADC Input Capacitance (C4, C6, C8)
    These capacitors are intended to filter high-frequency noise at the ADC inputs. Could this filtering negatively impact ADC performance?

  4. Analog Input Configuration
    I’ve attached a schematic showing the analog front end for both XLR and TRS inputs. The single-ended to differential converters are biased at 2.5 V (VCOM = 2.5 V) in a single-supply 5 V configuration. The outputs are AC-coupled to the ADC using 22 uF capacitors to remove the 2.5 V bias before entering the codec. Is this an appropriate approach?

I’d greatly appreciate any insights, suggestions for improvement, or corrections if there are issues with the design.

Thank you!

5076.schematic (2).pdf

  • +1
    TI__Mastermind 46425 points

    Hi Shemaiah,

    1) You're correct that the Bias Voltage is only required for microphone signals. There are some extra details around the exact biasing strategy depending on if you are AC coupling the input or not, but those are found in the datasheet. Edit: the answer to question 4 clarifies this

    2) Phantom power is the same as MICBIAS in this case. The XLR input will provide the signal along Pin 2 and 3, and the codec will provide MICBIAS back along Pin 2 and 3 to power the condenser mic. A TRS cable would be a similar idea.

    3) No, the capacitors will not hinder ADC performance.

    4) Yes you are doing the right thing. For AC coupling, the codec will provide its own bias. This allows the SE to DIFF converter to work correctly without the two bias voltages contesting.

    As another note, MICBIAS (or VBIAS in your schematic) is being provided after C1 and C5, so the 5V phantom power won't make it back to the microphone. It will be blocked by the caps. Moving the connection to the jack side should fix this.

    Best regards,
    Jeff McPherson

  • 0 in reply to Jeff McPherson
    Prodigy 20 points

    MICBIAS is provided after the capacitors as the intent is to have 48V phantom power to power condenser microphones and to bias that signal with MICBIAS. Is biasing not required here either?

    Edit: The application is a professional audio effects processor and this analog front end is intended for condenser/dynamic microphone input via the XLR connection and instrument input via the TRS connectors. 

    TAC5112 codec was chosen for this application as the internal programmable gain simplifies the hardware design and is footprint compatible with the higher performance TAC5212. However, the datasheet does not explicitly describe a PGA (programmable gain amplifier) or analog preamp block so is this application valid? 

    Additionally, the datasheet has "Professional audio mixer/control surface" as a recommended application however the linked webpage displays 404 error: https://www.ti.com/solution/professional-audio-mixer-control-surface.

    Below is a revised schmatic that removes the 4.22k resistor as internal 5k input impedance is present and the pull up to VBIAS. Would appreciate any feedback. 

    analog_input_v2.pdf

  • 0 in reply to Shemaiah Rangitaawa
    TI__Mastermind 46425 points

    Hi Shemaiah,

    Looking at the revised schematic, this is fine. The 48V phantom power will supply the mics, and so the VBIAS from the codec isn't required. The internal bias for the ADC is provided on the pins internally and is not related to MICBIAS.

    You're correct there is no analog gain stage in the TAC5112 (or TAC5212) but digital gain is available.

    Best regards,
    Jeff McPherson

  • 0 in reply to Jeff McPherson
    Prodigy 20 points

    Ok, thanks for the feedback and clarification.

    Would you recommend an analog microphone preamp (INA333/INA217/Op-amp-based or discrete) be placed before the ADC of the TAC5112 when using dynamic/condenser microphones or is digital gain sufficient, what are some considerations for this application? 

  • 0 in reply to Shemaiah Rangitaawa
    TI__Mastermind 46425 points

    Hi Shemaiah,

    It depends on the specs of your mic, or what kind of performance is expected. Digital gain won't improve the SNR in anyway, the noise will be amplified as well. Since you mentioned this is an effects processor and performance is critical, I'd recommend a pre-amp stage, similar to what you've done on the TRS side, just with a gain higher than 1. The full scale of the differential input to the codec is 2Vrms.

    Best regards,
    Jeff McPherson