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.

TLV320AIC3254: Beep generator ruins audio output

Part Number: TLV320AIC3254

Tool/software:

Hi all

Bit of a strange one. I've found that if I use the beep generator to make a short tone (in my case a short key beep) that afterwards the normal audio appears to have severe loss in medium to high frequencies. Using simple voice it sounds like massive roll off of all the good voice frequencies. I haven't measured it but I'd say everything above about 600 Hz has rolled off.

Is this sort of thing common? Does anything know of anything I can check to solve this?

Cheers

Matt

  • Hi Matthew,

    I've not heard of this happening before. Does the filtering effect still happen if you don't play audio during the beep? Maybe try the example found here if there's an issue in the configuration: https://e2e.ti.com/support/audio-group/audio/f/audio-forum/773086/faq-tlv320aic3254-tlv320aic32xx-family-beep-generator-configuration

    Best regards,
    Jeff McPherson

  • Hi Jeff

    Thank you for the quick response. I have just discovered that I need to disable the Beep power block after use and then the normal DAC audio still sounds good. So that problem is solved.

    My next issue is that the frequency generated by the beep generator doesn't seem to match my calculations. I made an excel sheet and used the sample numbers in the app note (which BTW lists 44.1Khz yet uses 48KHz), when I adjust for my 8 KHz I seem to be about 2.5 times off frequency. It is always 2.5 times out so I can deal with it but I'm keen to work out why / where I'm going wrong. I have 2.048MHz on MCLK, 256KHz on BCLK, 8KHz on WCLK. 

    For Fs = 8KHz, Fin = 1KHz, my calculated SIN = 0x5A82, and for COS = 0x5A82 I get a beep generator output of around 400Hz.

    To get closer to 1KHz (not exact) audio output I'm using SIN = 0x7641 and COS = 0xCF04

  • Hi Matthew,

    I double checked the equations, and you're right the beep length function uses 48kHz by mistake, but the sine and cos functions use 44.1kHz correctly.
    I also ended up with the same numbers as you did working through the equations. There's probably an error we need to work out internally, but you're following the equations correctly from what I can tell. If adding in the 2.5 factor works then I don't see a problem with it.

    Best regards,
    Jeff McPherson

  • Thanks Jeff. I can deal with the error, was just curious why really. No stress at all :)