Announcement

Collapse
No announcement yet.

Audio Level Variations during Playback

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Audio Level Variations during Playback

    I've noticed the audio level change during playback, specifically, getting louder.

    Does the Somo14D have some type of AGC going on?

    I've checked the audio source files very closely. The variation is coming from the module. The files being played are set as loud as possible, in that they are right at the clipping limit. The unit is running off a bench supply with plenty of current.

    It is driving an 8ohm 1/4w speaker directly.

    What could be causing this?

    Thanks.

  • #2


    Have you got at least 220uf as close to the 'chip' as possible?

    Try measuring the voltage right at the chip (i.e. both + and - of your multimeter n pins 8 and 9.

    I think you'll see the voltage fluctuating.
    Mark

    Comment


    • #3


      Dear ESP, thanks.

      I now see what it is, but left my camera at home so I can't include a snapshot till tomorrow. I'll send you a good scope picture of this.

      When the audio track starts to play, if you scope the speaker line, I see a voltage transient from 0v to 2V with a 1ms rise time, and a very slow exponential 500ms decay. The audio begins playing around 350ms, when the voltage is still quite offset.

      Also, the p-p level of the initial audio is very low compared to audio coming in at say the 1sec mark, maybe 5x less.

      My guess is that the audio amp is asleep, and this is a turn-on transient and settling time. If this condition can be avoided, I really need to know. In this application I was hoping to start audio with less than 50ms delay if possible.

      I'm also having trouble with it getting locked up. Maybe commands coming in while the thing is busy? I've been adding delays and watching the BUSY line, but it is still really touchy. The RESET sequence does not always revive it. I end up cycling power to clear it out when it happens. To get it to play, I send the filename. If I want to stop it, I send 0xFFFF and wait till BUSY drops.

      I checked for the voltage droop you mentioned, and only see a 150mV drop on VCC when the audio starts to play. It looks consistently flat too.

      Thanks,
      C



      ESPsupport wrote: Have you got at least 220uf as close to the 'chip' as possible?

      Try measuring the voltage right at the chip (i.e. both + and - of your multimeter n pins 8 and 9.

      I think you'll see the voltage fluctuating.

      Comment


      • #4


        snapshots showing a scope attached to one side of the speaker,

        ac coupled,
        dc coupled
        and the original audio waveform for comparison


        Attached files

        Comment

        Working...
        X