TPA6120A2: TPA6120A2 interface with the processor Board Codec ALC888S-VD

Part Number: TPA6120A2

Tool/software:

Hi,

Please find the Block Diagram for the circuit to help you understand how we connected the ALC888S-VD Board and TPA6120 to the Avionic headset, Bose A20. The gain we have set for TPA6120 is approximately 3 times the input. 
Please find our board circuit of the processor Codec i.e ALC888S-VD and TPA6120. 
TPA6120 and the processor module are routed with 1m of audio cable length. 
We captured waveforms with a scope and frequency response analyzer when we heard the noise, as attached to this email. Red is the TPA6120 Input, and Blue is the TPA6120 output.
Please suggest how to fix the noise at TPA6120 side. 
Thanks & Regards,
Jagadish.
  • Hi Jagadish,

    Thank you for providing the schematic and block diagram. I do have a couple questions. From looking at the schematics, I cannot find the inputs for the TPA6120 device on the codec schematic. I was wondering if there was an additional section of the schematic or if there is a renaming of that path. I was also curious if you could give some detail on the actual signal seen at HDPN_IN+ and HDPN__OUT, which I assume are your input signals. Are these signals offset to mid-supply? The concern being that in single-supply, if the inputs are not within the devices input common-mode range, you would likely not expect to see anything at the output as it is not in its linear region.

    Best regards,

    Ignacio

  • Hi Ignacio,

    Thanks for your response. The signals are connected between processor codec to TPA6120 is connected as follows.

    the signal of HDPN+ is connected to processor side J10.5 (A_R) and J10.6 (A_L). and HDPN- is connected to J10.10 pins. same shown block diagram for your reference

    already shared noise waveforms in the zip file. 

    Thanks & Regards,

    Jagadish.

  • Hi Jagadish,

    Thank you for the additional information. I believe the issue you are having is due to the input signal being out of the linear range of the device. The device requires the input signal to have an offset ideally around mid-supply which in this single supply configuration would be around 7.5V. The main signal would then ride on this offset. If possible, in the customers board, you could set the negative supply pins of the amplifier to -15V as well and the amplifier should behave as expected.

    Best Regards,

    Ignacio

  • we don't have negative supply in both the board and we are operating both the boards in singled ended instead of differential. 

  • Hi Jagadish,

    Understood, the issue seems to be the input being outside the input range of the device as was mentioned above. Another option you have is to ac couple your input and use pull up resistors to mid-supply at your two non-inverting inputs. You would also bias resistors R53 and R56 to mid-supply instead of ground. This would allow you to set your input-s to the appropriate voltage and have your signal ride on this voltage.

    Best regards,

    Ignacio

  • Hi Ignacio,

    could you please explain R53 and R56 to mid-supply means tie to +15V??

  • Hi Jagadish,

    I found a helpful app note showing two possible solutions for single supply configurations. The goal is setting the inputs to mid-supply. This can be done with a voltage divider like the app note shows to bias the inputs to mid-supply of 7.5V in this case. You could also use a single resistor (R53 and R56) directly tied to mid-supply of 7.5V instead of the voltage divider. 

    AC Coupled Single Supply Amplifier Reference Design

    Best Regards,

    Ignacio