With aid of the datasheet of the charger, I build up the underneath circuit with minor as follow:

  • The 1uH is replaced with a short circuit.
  • The 22uF capacitor is eliminated because we do not have any load other than the batteries.
  • The batteries does not have a build-in thermistor. Thus, an external thermistor is connected between NTC pin and ground.
  • The I2C pin is floating. I am not using I2C at all.


Our AC/DC adapter delivers 5V and it is rated for 9W loads. We have two batteries that are connected in parallel, each is 3.7V and 1.1 Ah. Below is a picture of the circuit that I built at home.

The Result

I need to know how much current is delivered to the batteries (because I do not have the right equipment for it). So, I rely on calculations to know the amount of the current delivered to the batteries. My calculations of the current are based on the change in batteries’ voltage over time. It sounds like the batteries are receiving 0.03mA only. This means, according to the datasheet, the charger is operating on a low mode.

In order to solve the problem, I did the following trials:

  • I connected CLPROG1 and CLPROG2 on two different resistors. This trial failed, meaning the charging rate still low.
  • I tried to resize the 1.21k Ohm resistor (see above picture). Again, I got the same result.


One of the possible errors is that our battery does not have a built-in thermistor because someone had this problem before:


Please, let us know what could be wrong with our circuit. Any idea to improve the design is appreciated.


  • 0
    •  Analog Employees 
    on Aug 7, 2019 2:52 PM


    Thank you for the information and for providing a photo.

    A breadboard prototype circuit is just not going to work for this part, most switching regulators operating at any reasonable power will have trouble if wired out like this. The reason is that everything is wired out with long distances and this adds parasitic inductance/resistance that will interfere with operation. For example, bypass caps cannot be located close to the IC as they should be, and the resistance in grounding between components will cause a ground bounce.

    For power ICs, really any switching regulator, you really need to just design it onto a PCB. Note that we carefully detail PCB layout instructions in the datasheet for these parts. For assistance with this, I recommend you reference the demo board layout since it is a working example.

    A breadboard circuit like this is perfectly acceptable for low-power circuits or digital components. It is just power circuitry that will have a problem.

    The NTC issue will not stop you from charging in general, so that isn't the issue here. Since your battery does not have an NTC, though, you might consider externally biasing the BAT pin with a current-limited voltage derived from the input. This will fix the operation that occurs when you use a fixed NTC hard-wired to the circuit.