• State
  • Locked Locked
  • Replies 5 replies
  • Subscribers 49 subscribers
  • Views 178 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

THS4551: VS+ to VS- short failure mode

John Vaughn
Prodigy 10 points
Part Number: THS4551

Tool/software:

Hi,

I have been working with a PCB design that integrates a THS4551 amplifier (DGK Package), and have been seeing occasional failures on this IC. The design will work as expected temporarily, but when it is fully integrated into our larger system, there have been cases of failures which, upon troubleshooting, show the THS4551 shorting VS+ to VS-, causing the entire design to fail due to our supply hitting its set current limit. There has beem no visible solder bridging, and the only way to fix this issue has been to replace the THS4551 in question.

There are many variables we are considering for how exactly this failure is happening, so my question to the forum is, what conditions would cause a THS4551 to show this VS+/VS- short failure mode? Is this characteristic of an ESD event, overvoltage somewhere on the IC, or perhaps something else? I am having issues finding forum entries discussing similar failure modes, so I'm hoping the support here can point me in the right direction for preventing these failures in the future.

Thanks,
JV

  • 0
    TI__Genius 10490 points

    Hi John,

    You are correct in the ways that this type of failure can occur. The most common case is an overstress condition on the device. It seems like the device works normally therefore we might be able to rule out any transient spikes occurring during startup. However, when the PCB is integrated into a larger system, the failures start to occur. Is there a way to check the pins on the device in the condition you are seeing failures? Any time even momentarily that a pin goes beyond the absolute max ratings specified in the datasheet there is a chance of catastrophic failure. Could you also share how many devices have been damaged and if this is happening on multiple PCBs? 

    Best Regards,

    Ignacio

  • 0 in reply to Ignacio Vazquez Lam1
    Prodigy 10 points

    Hi Ignacio,

    I am not able to check the pins in the condition where it fails at the moment, but I am working on getting a lab-friendly setup that closely emulates this condition. When that's available I'll be sure to closely monitor to see if anything pins are getting close to datasheet limits. I have seen this specific type of failure on ~5 PCBs out of a few dozen produced. 

    Does the failure mode as I describe it point to any specific pin(s) being overstressed?

  • 0 in reply to John Vaughn
    TI__Genius 10490 points

    Hi John,

    When replacing the device on the failed PCB does it occur again or is the PCB in the system behaving as expected once replaced? I would start with the supply pins for any voltage spikes on those pins in particular. If the failure occurs only when signals are running through the device, I will also recommend probing the input pins, however signs are pointing to an overstress at the supply pins.

    Best Regards,

    Ignacio

  • 0 in reply to Ignacio Vazquez Lam1
    Prodigy 10 points

    Hi Ignacio,

    I have not seen a repeat on the same IC after replacement yet. I'll be sure to look specifically at the supply pins. We are currently running on a Vs+ - Vs- of 5V, so close enough to the 5.5V limit that a transient spike may be in the danger zone.

    I am currently looking into the LMH6551Q as a replacement option with a higher max supply voltage, in case overdriving the supply pins is, in fact, the cause of our issues.

    Thanks,
    JV


  • 0 in reply to John Vaughn
    TI__Genius 10490 points

    Hi John,

    I would try retesting the PCBs that had failures and see if a replacement also fails as this might offer new insight as well. As for the pins to measure, the inputs as well as the supply are the most critical to measure for overstress. I will attach a PowerPoint that a member in TI has created that is worth looking over. It does a great job of highlighting major reasons for EOS failures and ways to protect the circuit depending on where you might be seeing these problems.

    2474.ESD_EOS (1).pptx

    Best Regards,

    Ignacio