• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 63 subscribers
  • Views 987 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ25895: Need Part Suggestion for Battery Charging by Solar or USB

Will Wang1
Prodigy 250 points
Part Number: BQ25895
Other Parts Discussed in Thread: , BQ25505, BQ25703A, BQ24650, BQ24253, BQ24250

Hello,

For my IoT application, I'm looking for a battery charger to support multiple power sources (solar cell and 5V USB).

  1. BQ25895
    1. I like all the features of BQ25895, but it's not for LiFePO4 based on the datasheet. Is there any work-around to use for a single LiFePO4 cell? Any other part suggestions?
    2. What are the differences between BQ25895 and BQ25895M?
    3. Do you have any concerns with the ICO function if multiple chargers share a single power source (wall AC/DC adapter) in star topology? For example, three nodes of 5W each and the source is rated for 15W. I found some valuable info about ICO in (http://www.ti.com/lit/an/slva812c/slva812c.pdf), but it's doesn't cover my case.
    4. My device has a single power input port that feeds directly into the charger VBUS pin (after input TVS protection circuit). User can plug either a solar cell (unregulated) or a wall adapter (regulated) as the source. Do you see any issues with this concept?
  2. BQ25505
    1. What's the maximum load current from the system? Assuming the battery is not charged enough for supplement mode.
  3. BQ25703A
    1. Support LiFePO4 and others, with lots of features
    Is this suitable for solar cell input? The part selection features "solar input", but I don't find how in the datasheet. I'm guessing some mix of ICO, V_INDPM, I_INDPM to implement MPPT algorithm? Any user guide to help? I'm happy with the F.OCV method.

Thank you,

Will

  • 0
    TI__Guru**** 166176 points

    Hi Will,

    In general, for simplest MPPT of a solar cell input, you need a charger that has adjustable VINDPM (feedback loop that regulates charger input voltage to prevent input source crash).

    Regarding BQ2589X with I2C controlled VINDPM and default 4.2V charge current, you would need host software disable charge (using GPIO connected to /CE pin) at startup.  The the host software must lower VBATREG to 3.6V and set VINDPM to your panel's MPP.  There is a reference design for your application at .BQ25895M has default VBATREG=4.35V instead of 4.2V like BQ25895.  Unfortunately, both have minimum VBATREG=3.8V which is too high for 1S LiFePO4.  Connecting multiple chargers in parallel that are  running in VINDPM is not recommended since this is equivalent to connecting multiple voltage regulation loops together.  They will fight and cause oscillations.

    Regarding BQ255XX, this family is for small panels and can only provide, at best, 100mA charge current.

    Regarding BQ25703A, it is similar in concept to the BQ25895, except higher input and wider battery voltages range.  If you have a host, I suggest this device and use a similar concept as was used for the BQ25895 MPPT reference design above.  If don't have a host, then I recommend BQ24650.

    Regards,
    Jeff

  • 0 in reply to Jeff F
    Prodigy 250 points

    Hi Jeff - Thanks for the quick response. I'd prefer a stand-alone solution if there is one to avoid waking up my host MCU for MPPT. It's all about lower power consumption.

    BQ24650:

    - Does it use F.OCV for MPPT like BQ255XX? or just a constant Vin threshold set by a resistor? I want to design so the solar cell still charges the battery even with a cloudy winter day.

    - If it's a constant setting, can I design so that the source can be either a 5V DC (from USB adapter) or a 5V solar cell (unregulated)?

    - Since this part doesn't implement power path topology, would it be able to power the main system with a drained or removed battery (assuming source is a USB wall adapter)? My main system sleeps 95% of the time, but it may draw 800mA at peak when awake.

    Thank you,

    Will

  • 0 in reply to Will Wang1
    TI__Guru**** 166176 points

    Hi Will,

    The BQ24650 has a fixed VINDPM treshold as a % of the current input voltage, not open circuit, voltage.  There is no sampling. 

    You can diode OR 5V supply and 5VOC, 4V MPP solar cell.  The charger simply limits the charge current if the voltage it senses at its input drops to that threshold, presumably ~4V for a 5Voc cell.

    Regarding power path, this IC does not have that feature.  I would recommend adding another diode from battery back to the input rail. 

    A better option might be the BQ24250 or BQ24253 with similar fixed, resistor settable VINDPM(MPP) threshold.

    I have an idea for creating a discrete sample and hold circuit, for sensing solar panel VOC, by using an SVS like as an oscillator. I built and tested an SVS as an oscillator but I have not implemented a full sample and hold circuit.  The VMPP as a % of that open circuit voltage would still be fixed.  But if you want to make it adjustable as a function of temperature, you can use the idea presented at the app note below, using series thermistors with the VINDPM/MPP resistors.  

    https://www.ti.com/lit/an/slua586a/slua586a.pdf

    Regards,

    Jeff