Is it safe to "bypass" (as in circumvent) a linear voltage regulator by applying a voltage to its output (equal to the specified regulator output voltage), while the regulator input remains unconnected?


I have a circuit which is powered by a 6V adapter (a "wall-wart"). The 6V adapter supplies the input of an ADP3338 (through a diode), which delivers 5V regulated output to the rest of the circuit.

Now I am interested to see how the circuit would respond to a different 5V supply. That is, I would like to power the circuit from an external 5V supply, bypassing the ADP3338 (without otherwise tampering with the circuit).

My idea was to simply connect an external 5V supply to the output of the ADP3338, while leaving the input (6V from adapter) unconnected. This is illustrated below:

regulator-part of the circuit

However, I am not sure if this is safe, or if it will destroy the regulator, or if it will mess with the rest of my circuit in some other way.

The functional block diagram from the ADP3338 specs, depicted below, shows a resistive voltage divider between GND and OUT. I measured 0.7M Ohms between these two pins, so I guess that should not be an issue. But how will the rest of the regulator circuitry react?

Thanks for your help.

adp3338 block diagram

  • 1
    \$\begingroup\$ if you connect a schottky diode from output to input of the regulator, this will limit the reverse voltage across the regulator to <0.3v, even if the regulator doesn't have such a diode within it. \$\endgroup\$
    – Neil_UK
    Mar 15 '19 at 8:49
  • \$\begingroup\$ @Neil_UK: thanks, I'll try that, just to be sure. \$\endgroup\$
    – djvg
    Mar 15 '19 at 9:15
  • \$\begingroup\$ @dim: Thanks, that question is indeed almost the same, except for the type of regulator. I cannot believe it did not turn up in my search... \$\endgroup\$
    – djvg
    Mar 15 '19 at 9:18
  • \$\begingroup\$ Another related question: electronics.stackexchange.com/q/252535 \$\endgroup\$
    – djvg
    Mar 15 '19 at 9:32

Although the datasheet doesn't mention it, you will find that most voltage regulators these days, have a protection diode between the output and the input, in order to allow for the capacitors on the regulated side to discharge. This also works if you were to apply an external voltage to the output of the regulator, and therefore it should not damage it.

As pointed out in the comments, there is no certainty about the presence of the internal diode, so to be on the safe side, you should add a Schottky diode between the output and the input.

  • \$\begingroup\$ Thanks. Is there any way, other than trial-and-error, to verify that the ADP3338 has such a protection diode? \$\endgroup\$
    – djvg
    Mar 15 '19 at 8:52
  • 4
    \$\begingroup\$ I would not see this as a correct answer. You should add a bit more to it. Yes, it is true that some regulators do have a protection diode, but not all of them do, so this answer is basically saying it's fine to apply an external voltage when in actual fact it is not. Unless you know 100% there is a protection diode, you should fit an external schottky to be sure. You should never assume things! \$\endgroup\$
    – MCG
    Mar 15 '19 at 8:55

Not the answer you're looking for? Browse other questions tagged or ask your own question.