• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 62 subscribers
  • Views 1926 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.

TPS25927: Catastrophic Failure when OUT shorted to IN?

Antenna_Head
Genius 11395 points
Part Number: TPS25927
Other Parts Discussed in Thread: TPS25940, TPS25942A

Working on a customer circuit that uses the TPS259270.  I imported the un-encrypted PSPICE model into TINA so a TINA native model wasn't available.

In the course of fault testing, the customer intentionally connects VIN to VOUT to simulate a specific short circuit fault condition on the board that might exist in real life.

Doing so, the TPS259270 is quickly and catastrophically vaporized.

Upon simulation, a dead short seems to appear on VOUT if VIN is connected to VOUT.  Here's my TINA schematic.  As this uses the unencrypted model, I can't post the TINA circuit here. R3 models the short from VIN to VOUT.

The simulation shows that when EN is low, OUT appears to be shorted to ground and an excessive current flows through IIN:

Notice IIN starts up at 110A, resulting in:

PLEASE CONFIRM THESE ARE THE EXPECTED RESULTS.  THIS ISN'T GOOD.

  • 0
    TI__Guru* 95911 points
    Hi
    Can you please share the customer schematic of TPS259270 device.
    How the customer is doing VIN to VOUT fault testing? Shorting through relay? what is the device condition during this testing (Enabled ?)?
    Can you elaborate more to analyze.

    Thanks
    Rakesh
  • 0 in reply to Rakesh
    TI__Genius 11395 points

    Rakesh-

    The TINA circuit included shows enough detail, and the simulation results are consistent with the destruction observed.

    Please comment or consult with the design team if, when EN is low, is VOUT shorted to ground?

    In the real application, this E-fuse is supposed to power an external device.

    It is possible the technician could inproperly wire up the external load and accidentally back-power the e-fuse with the same voltage that's on VIN.

    The product is designed to detect/report this condition, but obviously if the e-fuse won't survive we need a different solution.

    Thanks, Best, Steve

  • 0 in reply to Antenna_Head
    TI__Guru* 95911 points
    Hi Steve,

    For TPS25927 device, VOUT is not shorted to ground when disabled (EN= low).
    For the devices having "quick output discharge" feature, the VOUT is grounded through internal discharge resistor but not directly shorted.

    From your application description, I would recommend to chose back-to-back integrated FET device TPS25940/TPS25942A which can provide reverse current blocking to prevent "BACK-POWER". The FLT output of these devices can be used to report the condition.

    Thanks
    Rakesh
  • 0 in reply to Rakesh
    TI__Genius 11395 points
    Rakesh-

    Please explain why, when EN = 0, there is EXTREMELY HIGH CURRENT in the simulation and real life.

    The DATA SHEET makes no mention of what happens if the e-fuse is back biased when EN = 0.

    The block diagram implies that the substrate diode will simply allow current to flow from VOUT -> VIN, not to ground.

    We are not looking for another e-fuse at the moment.

    Please answer the questions I've asked- this is becoming urgent, and will need to escalate this to management if we can't get lucid answers through E2E.
  • 0 in reply to Antenna_Head
    TI__Guru* 95911 points
    Steve,

    Please feel free to setup call for closing this at the earliest and to support our customers.

    Regards,
    Rakesh
  • 0 in reply to Rakesh
    TI__Genius 11395 points

    Per Rakesh, there appears to be a problem with the SPICE model- hopefully the team can address.

    We will try to figure out a way to recreate this problem on the EVM.