• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 63 subscribers
  • Views 386 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.

LM2941C: Back powered LM2941SX behaviour

Curtis Figley
Prodigy 60 points
Part Number: LM2941C
Other Parts Discussed in Thread: LM2941

I am trying to determine the behaviour of LM2941SX LDO's if they are back powered from the output sides.

The particular situation I have is where there are two of these regulators, one set for 5V out and the other set for 12V out.
In the circuits these are driving, there are parasitic IC diodes in several mixed digital/analog chips that point from the digital 5V supply pins to the analog 12V supply pins.

If power from the 5V side is applied first and the input to the 12V side remains off, the parasitic diodes in the powered IC's will forward bias from the 5V side and try to drive the 12V supply pins to about 4.3V.

At this point, there will be about 4.3V applied to the LDO output when the input voltage is very nearly zero (in the circuit I'm looking at, the 5V input is always present, bu the 12V LDO input may not be energized or may actually be held at ground).

Are there any reverse body diode elements or other features that would allow significant current to flow into the output of the LDO if the input were either floating or held near ground through a low impedance?
For instance, is there a parasitic diode between the output and input?
Is there a mechanism where current would flow into the output and into the ground lead?

  • 0
    TI__Guru 55940 points
    Hi Curtis,

    I apologize for the delay in response with the US Holiday. This is an older design that was originally released by National Semiconductor before TI acquired the company. Please allow us some time to look into this.

    Very Respectfully,
    Ryan
  • 0 in reply to Ryan Eslinger
    TI__Guru 55940 points
    Hi Curtis,

    I apologize but we are still working on finding the information on this older design in order to confirm; however, from similar devices here is what I would expect your answers to be:

    For similar linear regulators there is no intentional path for reverse current within the device and any parasitic path would not have any current limiting circuitry; therefore, if significant reverse current is expected, we would recommend using external circuitry such as a diode. Since you have ample headroom (Vin - Vout), you could place a diode in series before the LDO input capacitor. The reverse current will charge the input capacitor raising the Vin rail local to the LM2941C until Vout no longer exceeds Vin so that reverse current will not flow.

    Very Respectfully,
    Ryan
  • 0 in reply to Ryan Eslinger
    Prodigy 60 points

    Hi:

    That would probably have been a smart thing to do before we had the boards in production......

    The LM2941C is advertised to have "Mirror Image Insertion Protection" which should mean they self protect and block backward current flow from the output to the input all on their own.
    I've seen a number of designs using the LM2941 for 12V lead acid battery chargers where these are used as the end of charge voltage limiter.
    They are typically left permanently connected to the battery terminals, even when the input supply is removed.
    There is circuitry on the input side of these designs that would drag the input rail down to some low value, so eventually most of the battery voltage would be impressed backwards across the LM2941.
    But nobody has indicated that this is an issue.


    In our case we're only applying 5V in the reverse direction, so notionally that should be less stressful than the 12V charger situation above.

    But just because people use them that way doesn't mean its a good idea....

    Do your old analogue guru's have anything to add?

    Cheers,

    Curtis

  • 0 in reply to Curtis Figley
    TI__Guru 55940 points

    Hi Curtis,

    I apologize for the delay.  We were able to confirm that there will be a leakage path from OUT to IN.  This path will not limit the current and is not intentionally designed for reverse current, characterized, or tested and is thus outside of the Recommended Operating Conditions.  

    The "Mirror Image Insertion Protection" is an older phrase for through hole packages that refers to the fact that the TO-220 package will survive being inserted backwards.  In this case OUT will be flipped with ADJ and IN will be flipped with ON/OFF- not OUT flipping with IN.

    Very Respectfully,

    Ryan