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.

TPS63900: Voltage tolerance and failure mode?

Part Number: TPS63900

Hi Folks

We've been trialling the TPS63900 on a test PCB as part of a wider test of an op amp circuit design.

We have two draft designs that would like to make use of the TPS63900, which was recommended by TI in response to a forum question, to support an ultra low power wireless design that has bursts of current when the radio is transmitting.  The low quiescent current,  current output capability, and dual selectable voltage of this new device would be ideal for the application.

During testing we had two over voltage incidents, both of which resulted in failure of the TPS63900.  One was due to a crappy power supply that took a moment to stabilise the voltage, so there was an over voltage for a short period.  When that happened the TPS63900 immediately shorted with magic smoke.

The second incident was when a normally decent lab power supply went haywire for a moment due to a high voltage electric fence test setup sparking to ground, causing it to briefly loose voltage control.  Clearly need to be more careful!

No magic smoke this time (the current control cut back in) but again the TPS63900 was immediately destroyed creating a permanent short.

My concern is twofold.   Generally the robustness of the TPS63900 to brief over voltage situations and secondly the apparent mode of failure being a dead short.

Are these issues recognised /, being addressed in a production version?

No chance to do formal testing of the device yet, so the 'evidence' is anecdotal, as above and not a very fair test, nevertheless doesn't inspire confidence in using the device in new designs.  Other parts on the test board survived.

Would appreciate your advice.

PS - We are still waiting on a TI response to an earlier question on availability of additional samples and production.

Cheers, Ron

  • Hi Ron:

    TPS63900 is a low voltage product, and the absolute maximum voltage is 5.9V. Stresses beyond those listed under Absolute Maximum Ratings may cause permanent damage to the device. 

    May you help to catch the waveform of the Vin, LX1, LX2, Vout to check if it do exceed the absolute maximum voltage. If so, the only thing we can do is modifying the surround circuit and avoid such overvoltage.

    And, may I know what's the earlier question? If for samples, you need to connect the filed team. 

  • Hi Minqiu

    Thanks for getting back. In practice, with battery operation, even though over voltages are very unlikely to occur, is there any specific protection circuit that you would recommend that wouldn't significantly impact current draw?

    The other question was about when the device is likely to be in production and also about further samples. So we are still waiting for an answer on this.

    Very best regards, Ron

  • Hi Ron:

    Firstly, you should take care the ESD in production, transform and shipment.

    Then try to avoid the over voltage case like hot swap. If the over voltage is inevitable, you can add TVS or Zener diode to reduce the peak.  

    For TPS63900, I remember it's sample available already. But I'm not sure when the it will in mass production. 

     

  • Hi Minqiu

    It's a pity that such a new apparently smart design doesn't provide protection by sensing an over-voltage condition and shutting down the converter. 

    We will have to implement protection if we are to use the converter with any kind of mains source power in case of power spikes.

    With regards samples, the web page is still saying 'not available'.

    Very best regards

    Ron

  • Hi Ron:

    Sorry, you mean the over voltage protect by itself? Yes, it will stop switching. But if the over voltage is forced by external circuit, it still has the risk to damage. 

    For the sample, you can contact the local filed team or distribution. We are different department.

  • Hi Minqiu

    You're right of course. Stopping switching won't prevent damage.   To keep it simple I'll need to try to find an ultra low leakage Zener.

    Have contacted 'customer support' again as they haven't provide an answer about production or samples since 31st July.

    Very best regards

    Ron

  • Hi Ron:

    It looks the device still need some time to release. Please wait with patience. Thank you!