USB Lithium Ion charger and Backup Power Supply circuit
A good power supply source is essential for microcontroller-based project. Having a Uninterruptible Power Supply backup circuit is essential for microcontroller projects where we need constant power source without any cut-out. This post describes the design of a simple USB Lithium Ion charger which forms part of a 3.3 V power supply circuit. Whenever mains supply fails, the lithium ion battery takes the load without any delay and when mains supply is restored, the battery goes to charging mode again.
Lithium Ion Charger Circuit Design
The core of this circuit is the MCP73812T IC from Microchip. This IC is a linear charge management controller that provides specific charge algorithm for single cell Li-ion batteries. The IC can be powered by a single 5V which makes it perfect for a USB powered charger. The charging current is set by a single external resistor. The circuit presented here is a USB-powered Lithium ion charger combined in a backup powered supply.
The circuit is powered at 5V through a micro USB connector. When the USB is powered, the FET Q1 is turned off and current flows to the Lithium Ion charger IC and to the 3.3V regulator (MCP1603). Under USB power, the Lithium Ion battery charges. When the USB power is removed, the FET Q1 turns on and current flows from the battery through the Drain-to-Source of Q1 into the 3.3V regulator MCP1603.
Components list
A 3D model of the circuit is shown below:
Assembled Circuit
The circuit was assembled and connected to a PC through a USB cable. The green LED indicates that the board is being powered by USB. The Lithium-ion battery was also connected and left to charge. A stable voltage of 3.3 V was measured at the circuit’s output. When the USB cable is disconnected, the LED goes off and the 3.3 V at the output is maintained.
If you have any question and/or suggestions, please contact us at contact@behindthesciences.com or leave a comment below 🙂
[1] https://www.olimex.com
Related Posts
About The Author
behindthesciences
We are two engineers who work in the Electronic and Telecommunication industry. We regularly blog in our spare time.
@Behind the Sciences. Nice. I’ve done something similar with the load switch. There is one catch, though, and whether or not it is a problem depends on the size of your battery. The Ir through D2 is constant when the circuit is running from a battery. The NXP diode shows this reverse bias current to be 55uA at 5V. If you are looking for a long-term battery life, this current could be a substantial drain. Some of the better MCUs have a low-power mode in the 1-5uA range. Ir on D2 is 10x that. And, it doesn’t matter what value is used for R3, the current is always going to flow.
The reverse leakage current is about 10uA for that diode. I agree that the leakage might be an issue but the purpose of this circuit is mainly to be as a UPS. Thanks for reading and commenting 🙂
Hey guys im pretty new at this but i have a few questions. The data sheet for the MCP1603 calls for a 4.7uF capacitor but your circuit is using 22uF is this just to get a better signal? Also as far the large reverse current on D2 is it just a matter of using a different diode or is there more to it. What would you recommend as a replacement?
We used a 22uF here as it filtered the noise better and reduced the ripple voltage better. You can of course select a diode with a lower reverse current.
What can you do to fix that problem?
Which problem are you referring to?
I have doubts whether the design works as it is described in the article. In my opinion, Q1 transistor should have a changed source with a drain. In current solution, reverse diode inside Q1 always conducts current from BATT to LDO. It does not matter what happens on the Q1 gate… You can replace Q1 with a Schottky diode, it would be the same effect. But you wanted to have a MOS-FET switch here, so something is wrong here…. I am waiting for your comments on this matter.
It will conduct a small leakage current only if there is a potential difference.
A schottky diode won’t do the same job here as it is evident what the purpose of the MOSFET is here. You are welcome to improve it. In our product, it did the job well and we tested the leakage currents, which was not an issue. 🙂