No-microcontroller soft latching power switch

Thread Starter

awwwt

Joined Aug 24, 2015
60
Hi everyone,

I am currently trying out a bunch of soft latching power switches, as I would like a push-on-hold-off mechanism, where holding a momentary pushbutton for about 2 seconds latches the circuit off, and a short press turns it on.

I have tried the types of circuit that are designed to be used with a microcontroller / Arduino, which work very well, specifically this one and this one, and now I am curious to try a switch that does not require a micro controller to turn the circuit off!

I tried EEV blog’s switch here however the oscillation I experienced from this circuit is slightly too buggy for me to implement as the circuit turns on again quickly after holding it off. Some of the comments on this video refer to oscillation being an issue and I couldn't find a fix for it without removing the hold-for-off 'feature'.

I have also come across Mosaic Industries’ circuit here, however I am looking for something with less complicated components (eg. simple through hole components like P and N channel fets and schotty or rectifier diodes, rather than smd chips and so on...)

My input voltage is 4.5v (3xAAA batteries). I was wondering if anyone had experience with these circuits and could point me in the right direction? Are these circuits when standalone generally quite difficult to achieve in a simple way when compared to a microcontroller-enabled one?

Thanks for reading this far!

Alice
 

jpanhalt

Joined Jan 18, 2008
9,420
You pretty much cover what I have in my folder for push on/push off switches. I tend not to take much advice directly from YouTube.

The LTC2955 is attractive, but may require a micro to get the long-push off function. The LM9074 (TI) seems to be an older design.

For a recent design, I chose a Maxim MAX16054. It is a simple push on/push off but requires no external components, except maybe for decoupling.
 

Thread Starter

awwwt

Joined Aug 24, 2015
60
You pretty much cover what I have in my folder for push on/push off switches. I tend not to take much advice directly from YouTube.

The LTC2955 is attractive, but may require a micro to get the long-push off function. The LM9074 (TI) seems to be an older design.

For a recent design, I chose a Maxim MAX16054. It is a simple push on/push off but requires no external components, except maybe for decoupling.
Thanks so much for this, super insightful! For the MAX16054 can you time the off so that it takes a long push, or is it a simpler toggle function?
 

MrChips

Joined Oct 2, 2009
21,126
Why do you want a long 2-second press to turn OFF?
This would appear to be not necessary.

If the unit is ON, the next press would turn it OFF.
If the unit is already OFF, why would you want to turn it OFF again?
 

Thread Starter

awwwt

Joined Aug 24, 2015
60
Why do you want a long 2-second press to turn OFF?
This would appear to be not necessary.

If the unit is ON, the next press would turn it OFF.
If the unit is already OFF, why would you want to turn it OFF again?
The reason why I'd like a longer press for OFF is so that in the interim I can use the button as an input signal to a microcontroller while it is not being used for power control separately. I would like the microcontroller to take the button's input but have nothing to do with its own powering ON/OFF - hence the need for a no-microcontroller solution for the press-on-hold-off function.
 

MrChips

Joined Oct 2, 2009
21,126
The reason why I'd like a longer press for OFF is so that in the interim I can use the button as an input signal to a microcontroller while it is not being used for power control separately. I would like the microcontroller to take the button's input but have nothing to do with its own powering ON/OFF - hence the need for a no-microcontroller solution for the press-on-hold-off function.
Ok. Why didn't you say that in the first place.
The simple solution is to create a 2-second timer using an RC circuit plus a diode for fast discharge.
 

crutschow

Joined Mar 14, 2008
24,968
Below is the LTspice simulation of a PB toggle circuit using CD40106 CMOS Schmitt-trigger inverters (one-half package), with no ON delay and a 2 second OFF delay (as determined by R3C3).
EDIT: A P-MOSFET can be added as a high-side power switch (since the MOSFET provides a signal inversion, connect R3 and D1 to the SW node so the delay still occurs for power OFF).
It will operate from 5V to 15V power, and draws only leakage power in the static state.

1587406234886.png
 
Last edited:

Thread Starter

awwwt

Joined Aug 24, 2015
60
Hi again, I have successfully managed to get the recommended MAX16054 working with my no-microcontroller circuit using a P-MOSFET such as in the attached diagram.

I am now wondering if the MAX16054 could be somehow bypassed or taken control of by a microontroller, where the same button is used to turn on/off the microcontroller, however with a long-push for off. In this case the original no-microcontroller circuit can be ignored.

To elaborate, the MAX16054 would switch a microcontroller on, then the microcontroller would keep itself on (possibly through the use of switching a different transistor with a GPIO pin?) So it could bypass any short button presses and also take the button as an input to a GPIO pin instead. It could then time the button's presses, and turn itself off again after a long-press (maybe by working with the CLEAR of the MAX16054?), thus resetting the system.

I am only just getting to grips with transistors and am not sure if the above would work at all, so any help would be much appreciated.

Thanks!
 

Attachments

Top