• State Resolved
  • Replies 4 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 73 views
  • Users 0 members are here
Support feedback
Options
Options
Related

LM5176-Q1: Overcurrent protection not suitable for failed short MOSFETS?

Joey van Hummel
Prodigy 50 points
Part Number: LM5176-Q1
Other Parts Discussed in Thread: LM5176

Tool/software:

Hi all,

In the LM5176-Q1 there is overcurrent protection. It senses the voltage across CS/CSG.

In our design this is used in combination with hiccup mode.

We have verified that these functions work in our design. e.g. when using a short at the output causes CS/CSG to rise and SS to be pulled down.

However, when testing what happens in buck-mode when HDRV1-FET is shorted, there is no cycle-by-cycle limitation performed and no hiccup mode entry.

My hypothesis is that CS/CSG in buck mode expects current from GND to the load. When HDRV1-FET is shorted, instead current flows from source to GND.

The polarity of CS/CSG voltage is reverse and thus there is no overcurrent detected. The threshold is never reached.

Is my reasoning correct? Or should LM5176-Q1 detect this fault also?

Kind regards,

Joey

  • 0
    TI__Guru 58641 points

    Hi Joey,

    thank you for using the E2E forum.

    You are right with your assumption. 

    In Buck mode the LM5176 uses Valley current mode and senses the current in the off-phase of the high side switch - as you described above.

    Best regards,

     Stefan

  • 0 in reply to Stefan Schauer
    Prodigy 50 points

    Hi Stefan,

    So just to clarify, in case of a high-side FET short, the short circuit current from source to GND in the off-phase would not be detected since the polarity is reversed?

    Thank you for your quick reply.

    Kind regards,

    Joey

  • +1 in reply to Joey van Hummel
    TI__Guru 58641 points

    Hi Joey,

    yes, this is right. It also will not prevent the pass through from VIN to VOUT (through the body diode of the Boost high side MOSFET.

    The current limit main purpose is to protect the power stage in case of overload or short circuit in the output. 

    It is very limited it protect on fails of the power stage circuit itself. For that another circuit part would be required, e.g. a fuse.

    Note: with an hard fail of a damaged MOSFET, it is also not possible that the circuit will recover again.

    Best regards,

     Stefan

  • 0 in reply to Stefan Schauer
    Prodigy 50 points

    Hi Stefan,

    Thanks again for the fast reply, much appreciated!

    Kind regards,

    Joey