I'm developing a module with integrated battery, where the power supply will came from a solar panel.

  • Solar panel is this one -- 5.5V - 90mA
  • Battery is this -- 3.7V - 1200mAh
  • My device is a Sensor with a Wifi transmitter. Consumes 3.3V - 30mA (300mA peak during 2 seconds, every 5 minutes)

According the battery seller, it comes with integrated IC to protect the battery.

  • Discharge cut-off voltage: 3 V

  • Charge cut-off Voltage: 4.2 V

Considering that, can I do my circuit simple like this one (without charger management)?

  • \$\begingroup\$ You can consider a battery charger IC which realize the proper CC/CV charging algorithm for charging LiPO cells. Such as this one, or there are more advanced one on the market. \$\endgroup\$ – Bence Kaulics Sep 1 '16 at 18:12

Could you? Sure. Is it recommended? Not at all. Slight risk of explosion. The battery's protection circuit will help with that, but without specifics of the IC and schematic used, we can't tell.

Many cheap solar powered lights do this very thing, directly charging a battery from a solar panel with no regulation. It's cheap, fast, and works okay. Cutting corners for a profit. Depending on your load and the sun that day, it may not charge all the way. At best, if we imagine 6 hours of solid light, that's only 90 mA * 6 h = 540 mAh. Not even half the battery capacity.

At the very least, you need a protection diode between the solar panel and battery, to prevent the battery from discharging into the solar cell.

And at these voltage and current, a linear regulator may be more efficient. You should look into that, as well as calculating in the IQ of the regulator and circuit.


At an abolute minimum, put a diode between the solar panel and the rest of the circuit, discharging into the solar panel will damage it. Because the maximum output of the solar cell is just 90mA (0.075 C), there is no risk of charging the battery at too high a current, though there is a risk of bringing the voltage too high. A simple LM317 set to 4.2V should be enough to make this work reasonably well. As the other answer pointed out, though, the solar panel will not be able to charge this fully in a day, nor will charge accumulate over several days. At 30mA, it will be fully discharged within, at an abolute maximum, 12 hours (assuming starting from fully discharged and charged only by the solar panel). Two such panels in parallel (with diodes from both) should give you the current that you need if you live in a particularly sunny area. If you can sleep the wifi sensor at lower current between the 5-minute intervals, then it becomes viable for this to run continuously.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.