• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 64 subscribers
  • Views 827 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.

TPS61130 failure risk analysis

Francesco Rizzo
Prodigy 30 points
Other Parts Discussed in Thread: TPS61130, TPS63051, TPS63020

Context:

We developed a wearable biosignal recorder, which uses TI's TPS61130 converter, with output voltages 3.6V (switching output) and 3.3V (LDO output, with 3.6V as input), and average load current about 40mA. The battery is a rechargeable Li-ion 3.7V. The TPS61130 is the only component that is able to steadily step up voltage on the board. Attached is the schematic's section of DC-DC converter as implemented in our board.

Question:

For an aerospatial application (in this case we will use a battery pack composed by a series of 2 primary Li cells 1.5V), we have been asked to make a risk analysis, in order to exclude the possibility that voltage, in "catastrophic" failure conditions too, can rise to values that can result in dangerous currents for the subject wearing the system. NASA indicates as voltage threshold 32V, beyond which a stronger safety analysis must be done.

From my initial analysis, I located a possible risk in failure of feedback divider resistances, that can result in increase of the output voltage and of the switching voltages on SWP/SWN. Which value can reach this increase, before damaging the IC? How much time can go on a voltage transient over nominal absolute ratings? Another risk could be the failure of the capacitor between inductances of SEPIC circuit: in my experience, an immediate overheating of the coil, subsequent to the increase of current, damages the insulations in few seconds and generates a short circuit between Vbat and GndPwr, so that there is no more voltage at the output. Can you see other possibly dangerous events? Do you have some test results or model simulations about these matters? The 32V threshold is far enough from the real values involved, or must I introduce additional safety solutions?

Thank you for your attention and help!

Best regards,

Francesco Rizzo

  • 0
    TI__Guru*** 148310 points

    The device will definitely fail and start shorting if you even get close to 32V.  This is well above the abs max ratings for the silicon and will surely break it.

    But I don't believe this is the best device for a new, low power portable application.  Since 2002, many new devices have been developed which over much smaller solution size and sometimes integrated OVP.  If you are using the SEPIC because your Vin goes above and below your Vout, then I would highly recommend the much smaller TPS63xxx family of buck-boost converters.  The TPS63051 is our newest and smallest one.  

  • 0 in reply to Chris Glaser
    Prodigy 30 points

    Thank you, Chris, for your answer and suggestions about new products.

    I'll try to better explain my question:
    while holding that the energy source comes only from a 3V battery pack, can I obtain from TI some information, coming from tests or simulation models, about what happens in critical situations (failure of components, abnormal operation...) at the output voltage of TPS61130?
    I have to demonstrate that, in the worst possible situation (catastrophic condition, multiple failures to the involved components) too, in any case there will not be an increase of voltage at the output or in other nets connected to the component, that could be dangerous (conventionally established at 32V for space applications).
    I'm quite sure that it's impossible to reach much higher voltages than declared as absolute maximum declared in datasheet, but I'd like to have more detailed values, coming from design analysis.


    Can you help me about it?

    Thank you very much!
    Best regards,

    Francesco Rizzo

  • 0 in reply to Francesco Rizzo
    TI__Guru*** 148310 points

    These kind of failure analysis (open/short tests they are sometimes called) are best done on your final system so that you can see what the impact to your particular system is.  Other customers have spreadsheets where the waveform resulting from failure conditions is recorded to show how the device responded.

    We can't put an exact number on where the device will fail or the highest voltage it can support, but typically this is no more than 10-12V.  

    I would again recommend a newer device with integrated OVP which would not fail under some conditions where this device would fail.  The TPS63020 is one such device with integrated OVP.

  • 0 in reply to Chris Glaser
    Prodigy 30 points

    Thank you Chris!

    OK, I'll try and do the open/short test on my system.

    Kind regards,

    Francesco Rizzo