• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 3 answers
  • Subscribers 62 subscribers
  • Views 105 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.

LP5907: Internal MOSFET Diode Question

Darren (FAE)
Expert 5483 points
Part Number: LP5907

Looking at Figure 7 in the DS, you see the internal MOSFET is shown as having the parasitic diode pointing from OUT to IN.

So if VOUT is ever higher than VIN by 0.3V, current flowing through this parasitic diode internal to the MOSFET process could damage the MOSFET and destroy the device.

Is that understanding correct?

Darren

  • 0
    TI__Mastermind 29180 points

    Hi Darren,

    Yes this is correct.  To protect the diode and internal MOSFET we recommend a reverse current in this scenario to be limited to no greater than 5% of maximum steady state current.  That would be 250mA * 5% = 12.5mA in this case.

    Thanks,

    Stephen

  • 0 in reply to Stephen Ziel
    TI__Expert 5483 points

    Hi Stephen,

    1) Where do you get the 5% number?

    2) I'm assuming the 12.5mA steady-state limit is to prevent diode heating / thermal runaway / destruction. But if the reverse current was no longer than 1ms, is there any possible number you could give as a rough estimate as to what the diode could handle? Like 125mA for 1ms shouldn't damage it...?

  • 0 in reply to Darren (FAE)
    TI__Mastermind 29180 points

    Hi Darren,

    5% is a common rule of thumb we use for maximum reverse current in linear regulators.  We give this same answer to all reverse current requests and questions, unless there is something specific in the datasheet regarding reverse current (such as the linear regulator protects against it).

    We do not have an energy profile available for the body diode so we cannot offer a specific energy requirement to maintain.

    Thanks,

    Stephen

  • 0 in reply to Stephen Ziel
    TI__Expert 5483 points

    Hi Stephen,

    Let me confirm...

    [1]

    The reason this is a common rule of thumb across linear regulators is:

    The LDO internal MOSFET design and fabrication process generate a parasitic "body" diode.
    For LDO's with higher current capacity, the MOSFET generally tends to be larger than LDO's with less current capacity.
    This "larger" or "smaller" internal MOSFET means the parasitic diode is also "larger" or "smaller"

    Thus, with a higher-current "larger" MOSFET LDO, the parasitic diode is larger and can pass more current.
    With a lower-current "smaller" MOSFET, the parasitic diode is smaller and can't pass as much current before heating.

    From the above, you can see how the 5% rule could be applied.
    The parasitic diode that exists as part of the manufacturing process is sized along with the MOSFET, so can support "as a rule of thumb", 5% of the max steady-state device current.

    Is that an accurate understanding?

    [2]

    If more reverse current flowed, what would happen? Is the temperature rise the issue? Or is there another issue...?

  • 0 in reply to Darren (FAE)
    TI__Mastermind 29180 points

    Hi Darren,

    To your questions.

    1. In general, this is a great way of thinking about this issue.  Keep in mind that some LDO's may not source very much current but may still have a large pass FET, to help reduce Rds_on and maintain a low dropout requirement. 

    2. During reverse current operation, all of your protection circuitry is incapable of protecting the LDO.  So thermal dissipation is a big issue, as the diode drop itself is much larger than the Rds_on * Iout of the internal pass FET.  So the device will dissipate more power and generate heat much more quickly, and thermal protection is not available in this operating mode. 

    The body diode is a parasitic device so it is not as well controlled as something like the Rds_on of the MOSFET.  So the tolerance of this voltage drop may vary.

    In cases where large amounts of reverse current can flow, we would be concerned about the current ratings of the bond wires.  If there was a substantial peak current, such as a short to ground on Vin and somehow Vout was providing current through the body diode, we would be concerned that the bond wires would fuse.  There is a Raytheon fusing calculator we use for instances where high peak currents must be evaluated, although this question only comes up once or twice a year.  And I cannot recall if it is discussed with reverse current, however the concern on bond wire fusing remains the same.

    Thanks,

    Stephen