What is the cheap and easy way of powering MCU from Li-Ion battery, with automatic 3.3V regulator disconnect while charging, to properly charge the battery? I've came up with something like this: 
Is it ok?
\$\begingroup\$
\$\endgroup\$
5
-
\$\begingroup\$ In addition to what you have, you might be able to get by with just putting Schottky diodes in between each of the outputs of the regulators and the MCU supply pins. \$\endgroup\$– tcrosleyCommented Jan 7, 2015 at 18:27
-
\$\begingroup\$ Conceptually it looks fine, but like tcrosley I would do something to make sure the regulators don't see issues with reverse voltage. \$\endgroup\$– I. WolfeCommented Jan 7, 2015 at 18:28
-
\$\begingroup\$ I'm scared that Schottky diode will have large forward voltage (0.3+). Also, I'm going to use LE33, and in the datasheet you can see example with 5V regulators (LE50) about how you can make yourself ehm.. "adjustable" psu just by turning on and off the inhibit pins. I think it's safe to apply voltage to output while in inhibit mode. jm.pl/karty/LE33.pdf \$\endgroup\$– FelIOxCommented Jan 7, 2015 at 19:37
-
\$\begingroup\$ Yeah, you would have to compensate for the diode, or the MCU would have to be ok with receiving 3V instead of 3.3V. So do you plan on putting the top regulator in inhibit mode when on battery power(maybe by using the output of the battery regulator on the inhibit pin)? It might be ok even if you don't, it's not always deadly to regulators to tie outputs together, but I'd still be ready for the possibility. \$\endgroup\$– I. WolfeCommented Jan 7, 2015 at 21:46
-
\$\begingroup\$ A MOSFET can be used in place of a diode as a low voltage drop path when on. | \$\endgroup\$– Russell McMahon ♦Commented Jan 7, 2015 at 23:58
Add a comment
|
1 Answer
\$\begingroup\$
\$\endgroup\$
5
One part that immediately comes to my mind is a MAX1555, which is a charge management chip for a single LiPo (or Li-Ion ) cell. It too is designed to operate directly from a 5V source, and even has a separate 5V USB input with a lower charge rate. The reason I thought of it is that it has an output designed to drive a LED when it is powered and the cell is found to be in need of charge. So when the cell is fully charged it will turn off the LED. You might be able to use that LED output to disable the regulator. I'd also like to point out that I've actually used that chip very recently in an application similar to yours. In my case I do have a user operated power switch, but I've found that the MAX1555 still charges the cell perfectly even with the switch on, because the chip works by sensing the cell voltage, and because the charge current it supplies is significantly more than the MCU. That may or may not be the case for you but it might open another option. In my case the only effect was to make my charge cycle, normally about an hour, take just a few minutes more.
-
\$\begingroup\$ Yes! Thanks, I was planning to use mcp73831, but I heard it doesn't like load sharing. \$\endgroup\$– FelIOxCommented Jan 7, 2015 at 19:31
-
\$\begingroup\$ Also, It'll be better for me to power my circuit from usb rather than Li-ion after fully charged \$\endgroup\$– FelIOxCommented Jan 7, 2015 at 19:39
-
\$\begingroup\$ Now, if you are like me, the next thing on your mind will be how to CHEAPLY protect your Li cell from under-voltage. You know, if you let them go below about 2.8V, they will no longer hold a charge as well after that. Li cells are wonderful, but can also become a PIA. :-) \$\endgroup\$– RandyCommented Jan 7, 2015 at 19:55
-
\$\begingroup\$ Hmm.. I think batteries that you can get on Spark Fun have internal Undervoltage protection :) \$\endgroup\$– FelIOxCommented Jan 8, 2015 at 19:18
-
\$\begingroup\$ Please let me know of any you find like that. It might save me from having to update one of my protects, for about the 100th time. Others may be interested too. \$\endgroup\$– RandyCommented Jan 9, 2015 at 1:09