Post Go back to editing

Pop sound

Hi all,
In my project, I use an ADAU1701 with 3 outputs controlled by a MCU. In the path of both output 1 and 2 (DAC0 and DAC1), I've put a Voltage Controlled Delay. I've notices that, when I change the DC Input Entry level, there sometimes occur pops. These pops can be very loud and even occur on another output (also DAC2), so not only the one with the delay in the path. Also when I use sigmaStudio to control the I2C, this happens, so I guess the I2C communication can not be the problem. Any suggestions?
I've attached a scope image. The yellow line is the output of the amplifier connected to DAC2 (no delay in the audio path). The I2C instruction changes the "DC" which controls the delay in the path of DAC0, this is not supposed to have any influence on the audio path of DAC2.
Thanks in advance!
Best regards,
Guy
  • Guy,

    Any jump in the delay value used by the voltage controlled delay will likely result in a pop on your audio output. By changing the delay value suddenly, you're creating a discontinuous output that will leave either a hole in the output (increasing the delay) or an overlap of the output paths (decreasing the delay). The recommended use of the variable delay block is for it to be controlled with a smoothly-modulating signal, like an oscillator or a dc value slewed with a volume control.

    You may be seeing the pop on your DAC2 output because of the sudden discontinuity in the data memory. The DAC2 signal path still uses the common data memory, so changing the use of this data memory elsewhere in the signal path could still have an effect.

  • Hi again,

    I have attached a small block diagram. If I understood well, it's possible and normal that, when the value of DC_DelayL of DC_DelayR is changed, there is a pop sound at the output of DAC2, right? Is there no way to avoid this?

    Thanks in advance!

    Best regards,

    Guy

  • You should use safeload to load your delay paramters to the SigmaDSP. This will ensure that each audio frame is executed with a single parameter, and not half of the old parameter and half of the new parameter. I would also recommend not changing the delay by more than 1 lsb at a time. This will ensure a small step size with less chance of an audible click when the parameter is updated.

  • This question has been assumed as answered either offline via email or with a multi-part answer. This question has now been closed out. If you have an inquiry related to this topic please post a new question in the applicable product forum.

    Thank you,
    EZ Admin