I have a circuit that does not work as I intended/expected. It's been over a year since I worked on it, but recently Farnell send me the FET I originally intended to use but became "out of stock" for a year, so I thought about picking it up again. I've been learning electronics on my own, and I remember I couldn't figure out what was wrong. Since I don't really have anyone I can ask for help, I'll try here... For now I would like to get some feedback whether my idea is even just correct/feasible.
In the past, I've designed a led ring for my stereoscope with functionality that I wanted to have but couldn't find in commercial products (and I just wanted to do it). My previous version works, but if I forgot to switch it off, the lipo battery would slowly drain until it was discharged enough that the smps couldn't continue regulating. Not good for the battery (and simply bad design). So I started revision 2. Please note I do this mostly for learning. I could simplify things a lot by using a simple on/off switch but that feels like cheating
.
So I came up with this (only power section shown):
On the left an USB plug with a LiPo charger circuit. The two connectors on top (BAT/BAT_HEADER) go to the battery. On the right is the power supply to get 3.3V. Both sections work as intended (used them in the past and on the previous version). The yellow lines going to the blue bus on the bottom got to the PIC ADC and two GPIO pins.
My idea is the following: The FET separates the battery ground from Vss. So the battery doesn't drain. By pressing the startup button, the battery ground is shorted to Vss, allowing the circuit to start up. The PIC24 can then start up and control the FET to tie the two grounds together. The circuit runs as though it is directly connected to the battery. After a while (timeout) or when the user presses the shutdown button or if the battery becomes too drained, the FET is released and the grounds are separated again. No more discharging.
As far as I recall, I populated the power section and the thing worked as I hoped it would. As soon as I populated the PIC (IIRC) the circuit wouldn't start up anymore. So, before I dive in again, could someone tell me if my reasoning above is correct (although I do realize there are certainly better ways to do it). Any thoughts?