Welcome to the Battery Management Section of the TI E2E Support Community. Ask questions, share knowledge, explore ideas, and help solve problems with fellow engineers. To post a question, click on the forum tab then "New Post".
Q: bqSWITCHER charge control
I am looking at the BQ24100 and need the ability of the local processor to control the actual charger current. I will determine required max current in the processor by monitoring other things in the system, and will need at least 50mA steps in the control (max current is 1.5A). In the past they have used programmable resistors or subsituted a voltage source driving a resistor into the equivalent of the Iset pin, but have not done so with this part.
Are there other ways to implement charger current control? A digital pot is too expensive. Would a voltage driving a resistor into Iset work, or are there other cheap methods?
I was also thinking of suggesting a part such as the BQ24113, which is system-controlled.
Q: Re: bqSWITCHER charge control
The programmable resistors or subsituted a voltage source driving a ressitor into the Iset pin should work on this part. Also, a voltage driving a resistor into Iset woks. I think it is cheapest methods.
Even system-controlled (host control) IC still need to do same work to achieve that charge current control.
There is another option: using SMBus charger IC (bq24745 or bq24765). But the host also need support SMBus. So, this solution may cost more.
Depending on how large of a min/max charge current swing you need, you could use some resistor(s) to set a static value resistance for I_set1 and I_set2,in conjunction with a MOSFET (or BJT, for example) and drive this FET throughout its linear region - essentially using it like a variable resistor (that is, assuming you have spec'd out and found the appropriate device to handle the min/max in/outputs characteristics required for your application).
You could define how many steps (i.e. resolution) you need for a range of min/max charge current; then using a microcontroller's ADC channel's output (or even use a PWM output signal) to bias the FET where you need it to operate. That would work with the bq241xx chargers -- well actually I can speak for the 3 specific devices I have use in the past in a few of my designs (the xx = 03, 05, and 15), which were all pretty robust and reliable.
Basically, you could try a setup like this: have 2 resistors in parallel (for each I_set input), and the FET would go in series with one of these resistors. Then bias the FET like I mentioned above, assuming you chose the "right" FET (i.e. that's capable of operating safely within the range you need).