• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Answers 1 answer
  • Subscribers 31 subscribers
  • Views 450 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.

TAS5421-Q1: Overcurrent Behavior

Chris Bevard
Prodigy 10 points

Part Number: TAS5421-Q1

I have designed in the TAS5421 amp to drive a specially wound coil whose DCR is just over 3 ohms.  The input to the amp is being fed by a DSP PWM output operating in full symmetry.  There are multi-pole filters between the DSP and the amp to give a more sinusoidal input.  Gain is set to 20dB. What I am seeing is the IC will often assert the fault pin when the signal is initiated.  Per I2C an overcurrent fault is recorded.  Other times it operates without issue.  A high-frequency current probe on the output shows the max current detected is less than 500mA, nowhere near the 3.5A threshold.  Any guidance you can provide would be useful, but 2 specific questions:

1. What does an input over the 1Vrms recommended max result in?  Could an input slightly greater cause the fault detection circuits to falsely trigger?

2. The datasheet mentions a "diagnostic cycle" for the fault detection.  What triggers this cycle? How long before a fault will clear?  If I can remove the failure entirely, that is preferred, but if I know how to deal with it, that will be satisfactory.

  • 0
    TI__Mastermind 45035 points

    Hi Chris,

    I will answer your questions first, then comment on first paragraph.

    1. The maximum input voltage is due to the limitations of the power supply.  There is simply no more headroom on the input circuits beyond the 1Vrms max.  Knowing this, you could possibly damage the input stage if it is overdriven.  This should not trigger an over current fault event.
    2. When a fault occurs, the output stage is placed in Hi-z mode and the fault is indicated in the I2C fault register.  Then a Load Diagnostic is run to test the load.  If there is a load fault, a 860ms timer starts.  After it times out another load diagnostic runs.  If the fault still exists, the process starts over again.  Contrary to this, if the load diagnostics does not find a load fault, the output stage is placed back into play mode.

    What is the impedance of the coil at 50Hz?  This will be the load the amplifier sees.  The 3 ohms is slightly below our minimum load.  If the signal is low, it should not be a problem.  The turn on fault characteristic can be from the magnetization current of the inductor plus the capacitor to ground.  To test this you can remove the capacitor and see if the fault is eliminated.  The solution can be to find an inductor with a lower Q, such as a metal alloy type or reduce the capacitor value in the output LC filter.

    Regards,
    Gregg Scott

  • 0 in reply to Gregg Scott
    Prodigy 10 points

    Thank you sir.  Interesting you mention 3ohms being below minimum.  Spec shows 4ohm nominal and a range of 2-16 ohms.  Any way, good call regarding the output filtering.  I have no way to soft-start the drive signal but I can increase the output impedance since filtering is not as important.