• State
  • Locked Locked
  • Replies 8 replies
  • Subscribers 30 subscribers
  • Views 680 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.

How am I destroying my TAS5756?

Joe454
Intellectual 580 points


I've been burning up a few of these chips somehow, and am uncertain why.  They are used exactly as in the datasheet (mono mode), and I'm not driving them terribly hard.  But during testing, from time to time, the amp just fails, shorting both outputs to ground.  PVDD is 24V, running at 192kHz.

Our load is a little unusual but well within the published specs.  

Since there is DC, overvoltage, and thermal protection, what condition could get around them, and cause a failure?  Maybe a spurious discontinuity in the input signal?  Any other oddball load characteristic it's sensitive to?  

It works sometimes, but not reliably, so hints in tracking this down would be helpful.

  • 0
    Expert 1045 points
    Can you post your PCB layout?
  • 0 in reply to Jesse van der Zouw
    Intellectual 580 points

    I'll seek permission to do so, but it's effectively identical to the EVB. I'm pretty confident it's not a thermal problem, as the chip isn't getting (that) hot pre-failure.  Are there other aspects of the layout that could be troublesome?

  • 0 in reply to Joe454
    Expert 1045 points
    You mention your load being unusual, could you elaborate? Class-D does favor inductive/resisitive loads and can have trouble with capacitive loads (like piezo tweeters).

    Does your problem also occur when you do nothing? Otherwise it might be a problem with software integration.
    Does the chip let out smoke when it fails? Does it draw any power after it 'fails'?
  • 0 in reply to Jesse van der Zouw
    Intellectual 580 points
    The load is in fact a piezo tweeter; it's largely a scientific application, rather than for audio reproduction. But the load was carefully matched to be well within spec (and is almost purely resistive near the active frequency), and we've had good success with several long-term tests.

    But what is it, specifically, about capacitive loads that is troublesome? Would those somehow be able to cook the amp and bypass the usual protections?

    When it fails it draws a ton of power; if the current were unlimited I'm sure it would smoke. Both outputs short to ground at about 1 ohm. I'm not sure what you mean by "nothing" - if it's on and muted (pre-failure) it can probably sit indefinitely.
  • 0 in reply to Joe454
    Expert 1045 points
    I would put a resistor in series with the tweeter and see if that helps
  • 0 in reply to Jesse van der Zouw
    Intellectual 580 points
    Thanks. Why would this be expected to help?
  • 0 in reply to Joe454
    Expert 1045 points

    From Maxim's  "Amplifier Considerations in Ceramic Speaker Applications":

    "Speaker manufacturers always recommend a fixed resistance (RL) in series with the ceramic speaker, as shown in Figure 4. This resistor acts to limit the amplifier's current output when the signal contains a great deal of high-frequency content. In some applications, fixed resistance may not be needed if the frequency response of the audio signal passed to the speaker can be bandwidth limited. This would ensure that the speaker does not look like a short circuit to the amplifier"

    Please note that the resistor will form a RC highpass with the piezo, if you choose the value right it can work in your favor by cutting out unwanted frequencies.

  • 0 in reply to Jesse van der Zouw
    Intellectual 580 points

    Thanks.  The load has a significant real part, and is at minimum 4 ohms, regardless of frequency, so there should be no over current issue.  Plus, wouldn't this be expected to trigger the TAS' current protection?  (I'll at least experiment.)

    [ And don't you mean that the RC becomes a lowpass? ]

    I'd love to hear from TI on this ... are you there TI?