Join the Conversation


  1. For those who can’t tell, this is a *terrible* battery charger design. Battery chargers can be built using discrete components, but the number of things done wrong here is significant.

    Using a green LED as a voltage reference is questionable, but then simply doubling it without a trimmer for fine adjustment is worse. For current limiting, the best I can tell is that it depends on the host USB port’s protection, which won’t work out if this happens to be plugged into a 2A+ wall wart USB charger. The lack of protection for the case where there is a battery plugged in but no power source is abuse to your pass transistor that you generally want to avoid, since it is the only thing keeping the battery from being dead short connected to 5V.

    This kind of battery topology is almost acceptable for more robust cells, such as flooded lead acid, but for any kind of sealed cell, and particularly li-ion, is terrifying. Please never build this charger.

    1. Actually, Kenneth’s post should be ignored, he’s wrong. Green LED isn’t being used as a trimmer. I reviewed the schematic. It’s an isolated LED, in parallel to the feedback loop on the second Op-Amp, so has no bearing on the circuit operation. His remark about lacking current protection is also incorrect. No such protection is needed, as current does not independently operate apart from voltage. It does not matter how much potential current is available, it isn’t going to go through the circuit unless sufficient voltage exists to overcome the impedance of the components themselves. Lastly, if a battery is plugged in, but no USB is connected, there is no direct path to ground from the battery. At least 4 resistors are in place to prevent that. This is a cleverly designed circuit.

      1. Dear Moderator– please cancel my previous comment– there is a problem with the current level– I just didn’t see it in time. The transistor with its collector attached to the battery + terminal has no component to limit current into the battery.

        There should be a resistor on the ground lead to the battery to prevent a direct path to ground through the battery. That will fix this circuit.

        Warm regards

  2. I second that. Li-ion charger ICs are really cheap these days (<USD1 per piece) and there's a reason why they specify end-of-charge voltage to less than 1%. I too would be terrified to use this circuit.

  3. Thanks for the heads up, would avoid this circuit for Li-Ion now I see how it works.
    It might be OK for charging FePO4 cells for low current (ie solar) applications where they are happy with 3.65V to 10% capacity in mAh then shut down.

  4. I was wondering to know how do you design these resistor value ? pls tell me why , thanks a lot

Leave a comment

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.