Disable LDO from microcontroller

[metRo_]

Hello,

I'm going to use a TPS706 LDO on my circuit:

Could I use the EN input to enable it from a pushbutton and then disable it from the uC as I designed? +BATT will vary from 6.5V to 4V.

EDIT:
I just figured out that as soon as I shutdown the LDO the microcrontroller IO will be floating and the LDO will turn on again 

EDIT2:



Do you think that way it will work?

[Ian.M]

Yes, that can work, but when the button is pressed, you could get up to 260uA flowing through your MCU's input protection diode into the 3.3V rail.  That could cause problems for the MCU.   A Schottky clamp from the pin to the 3.3V rail would be advisable, which would also help satisfy the 0.4V max EN pin low threshold when the regulator output is disabled.

It gets more complex if you also want to be able to detect if the button is pressed while the MCU is running, and even more complex if you want to do that without using an extra I/O pin.

[metRo_]

Thank you!

It gets more complex if you also want to be able to detect if the button is pressed while the MCU is running, and even more complex if you want to do that without using an extra I/O pin.

I don't need to detect if the button is pressed.

A Schottky clamp from the pin to the 3.3V rail would be advisable, which would also help satisfy the 0.4V max EN pin low threshold when the regulator output is disabled.

When the regulator is disabled the microcontroller is also disabled, how could it not satisfy the 0.4V?

[Ian.M]

It only takes a few uA to develop 0.4V across a silicon PN junction of the area of a typical MCU protection diode.  For comparison, a 1N4148 rated at 200mA, will need about 20uA If to develop that Vf.   If due to contamination the leakage resistance across the switch drops below 3.5Meg it will no longer meet the EN low threshold.   

You are also going to have problems holding the pin low enough as the 3.3V rail drops and the MCU resets - you'll probably need to add a capacitor from the 3.3V rail to the pin end of the 10K resistor with a 100R resistor between that point and the pin.

[StillTrying]

The TPS706's EN switching levels seem to be 0.9V for a guaranteed On, and 0.4V for a guaranteed Off, and the current only 300nA, so I can't see the problem with the 10k and 2 Si diodes, maybe a 10 to 22nF across the 10k so the EN state doesn't change while the uP does its power up and down.