I'm trying to use a MCP73843 charge controller (ww1.microchip.com/downloads/en/DeviceDoc/21823D.pdf) to charge a 3.7V LiPo battery (4.2V max) off a 5V supply. With the exception of the optional diode, I built my circuit exactly per the application notes:


However, when I apply 5V (with my cell voltage at 3.8V), the same things always happens:

1. The LED is off for maybe half a second, with 65 mA drawn from power supply
2. The LED turns on and the current jumps to 71 mA
3. A half second later, the current jumps to 110 mA, the LED dims slightly, and then suddenly turns off (current drops to 0.23 mA)
4. Steps 1-3 repeat

I'm really baffled by this. In the datasheet, 1 Hz blinking of the status LED is supposed to indicate a "Safety timer fault", meaning it took more than 1.5 hours to charge the battery. But I reach this state right away! (even though I've checked the soldering of the capacitor and all the IC pins).

Anyone experienced with this particular IC have any thoughts?

 

I don't have experience with that IC, but I note that you are using a 100mΩ sense resistor. If I'm reading the datasheet correctly that implies a ~1A charging current (since it is stated that 220Ω gives 500mA). Can your cell and 5V supply both handle 1A?
Another thought: 100mΩ isn't much, so any additional resistance, e.g. due to a dry joint or other poor connection, would drop more voltage and cause the IC to pull down the charging current.

 

With the diode your supply is 4.3

 

Thanks for the replies! My cell and supply are both rated for 1A.

I ended up replacing all the components with new ones and now it works, lol. Maybe I was too heavy-handed with the soldering heatgun the first time around.

EDIT: Actually, now that I know it can work, I went and retested the old board (still had it) with thicker wires leading to my 5V supply. Now it works too. Guess the problem was indeed too much resistance!