• State Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 62 subscribers
  • Views 167 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

LMR51603: Reverse diode leakage connected to VIN

Don Herres
Prodigy 20 points
Part Number: LMR51603

Tool/software:

I am using an LMR51603XDBVR with input diodes to VIN.  This allows either of two power sources, nominal 48V, to power the circuit.  Two diodes with anodes to the voltage supply and cathode to the regulator.

One has a voltage divider in parallel to the diode input to a 110kΩ voltage divider to an input to a logic gate (48V -> 100kΩ -> gate logic at 4.4V -> 10kΩ -> GND).  The problem I have is the diode has reverse leakage when the other power supply is on and the input for this side is open.  The 110kΩ should pull it to ground, but I find about 45V at the diode anode.

This condition exists with this regulator, but not with either an AP7387 or ZXTR2105 linear regulator.  Nominal load is 20mA and neither of those will support that at 48V (ignore the AP7387 data sheet, that is not accurate).

I have tried increasing the 5V load from 20 mA to 50mA & 100mA, but there is no change.  The diode is a Schottky BAS70-00, but I had similar results with a BAS16W.

The circuit is from Figure 8.1 of the LMR51603 data sheet.  I have tried a 4.7µF cap for CIN with no change.

I also tried a 1µH inductor in series with the diode with no change either.  The problem exists with lower input at 42V & 32V.

  • 0
    TI__Expert 4900 points

    Hi Don,

    Could you share a schematic of your design?

    Thanks,

    Dino Maslic

  • 0 in reply to Dino Maslic
    Prodigy 20 points

    Attached is the portion of the circuit.  With 48V on input #1, I get about 45V on the D3 R6 junction with the input #2 open circuit.  This is isolated by lifting a pin on D3.  I tried reducing the resistors R6 & R4 with no significant change.  Also, eliminating D4 does not matter.

  • 0 in reply to Don Herres
    TI__Expert 4900 points

    Hi Don,

    This seems to be an issue with the diodes you are using and not the LMR51603.

    Since the issue is not with the WVB buck converter, I cannot comment on your protection diodes used.

    Thanks,

    Dino Maslic

  • 0 in reply to Dino Maslic
    Prodigy 20 points

    This, like most really interesting problems, is a system issue.  The diodes work correctly with a linear regulator.

    There are voltage transients at turn on and turn off that exceed the 75V rating of the diodes.  While not failing in a diode test, the leakage becomes worse after multiple cycles.  The transients are not consistent, so it requires multiple observations.  They are also dependent on the system inductance on the power supply.

    With a 0.1µF in parallel with diode D3 and an 85V diode, the problem is resolved because the transients are damped with the capacitor and the higher rating in addition.  I changed to 100V minimum components on D3, C3 and other components on the input power line.

  • 0 in reply to Don Herres
    TI__Expert 4900 points

    Hi Don,

    As you said, this seems to be an issue with the diodes you are using and not the LMR51603.

    Since the issue is not with the WVB buck converter, I cannot comment on your protection diodes used.

    Thanks,

    Dino Maslic