• State TI Thinks Resolved
  • Locked Locked
  • Replies 5 replies
  • Answers 2 answers
  • Subscribers 64 subscribers
  • Views 1192 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.

BQ24103A: Usage with solar panels

Gil Winograd
Prodigy 50 points
Part Number: BQ24103A
Other Parts Discussed in Thread: BQ24103, BQ25703A, BQ25600, BQ24715, BQ24650, BQ25703

For our battery charger, we are considering some part such as bq24103, which we have used on a different project. I wonder how well this interacts with a solar panel input.

 According to the data sheet, when the charger is in CC (constant current) mode, it will demand a fixed current determined by the setting, such as 1.5A. Now if the solar cannot provide 1.5A, then I would expect the voltage on the solar panel will collapse below the battery voltage, at which point the charger will enter a sleep-state. This will allow the panel to re-establish some voltage, but then the charger will turn back on and demand 1.5A again, in a cycle.

However, we actually tried the experiment and found that if the panel can only deliver a small current, such as 100mA, then the system will settle into a steady state where the charger is delivering 100mA to the battery, (which is in fact a desired behavior). This doesn’t seem to be documented by the datasheet, so is there an explanation for it ? Would this be typical of all similar chargers with integrated DC/DC, or will this behavior depend on the charger, and in that case, how can we know the behavior in advance from the datasheet ?

 

Thanks - Gil

 

  • 0
    TI__Expert 8690 points
    Hi Gil,

    Most likely the solar panel voltage is hovering above the sleep comparator threshold.

    I don't recommend running the solar panel in an open loop, it is better if the device has an input voltage regulation loop (in the chargers we call it VINDPM regulation, see bq25703A or bq25600, you can also search in our TI website for the filter Vin DPM (OK for Solar Power Source)). This will allow you to set the charger to regulate the solar panel in its Maximum Power Point. This should be available in the solar panel's datasheet. If not, usually the Maximum Power Point is 80% of the open voltage (no load).

    It would be useful to see the SW, AVCC, PVCC and VBAT waveforms.

    Regards,
    Steven
  • 0 in reply to Steven Bangdiwala
    Prodigy 50 points

    Thanks very much for your reply.

    We have decided to use the 2-series-cell, so I think that rules out bq25600.

    The bq25703A looks very attractive. I do have one question about the VDPM strategy... it seems the VDPM will reduce the charging current to maintain VIN at MPP, however, there is no indication that it will reduce the system-load current in the power-path as well. It seems to me that if we really want to maintain MPP in low-light even when there is some system load, then shouldn't we be drawing the system load from the battery ? Does it make sense for us to simply connect our load directly to the battery instead of using the power-path ?

    I was also looking at bq24715, this charger is advertised on the TI website with the Vin DPM (OK for Solar) feature, but I couldn't find anything about it at all in the actual datasheet.

    The last charger I was looking at is bq24650. However, the datasheet here shows I(AC)=25mA, this seems huge and pretty much rules out operation in low light. Does the bq24650 have any lower I(AC) current if the charging current becomes small due to VIN regulation ? Also, the recommended circuit shows a diode directly after the cell, won't this reduce the efficiency ?

    Thanks - Gil

  • 0 in reply to Gil Winograd
    TI__Expert 8690 points
    Hi Gil,

    The bq25703A has battery supplement, so if the input voltage limit is reached, then the battery will help with the system load. The input will always stay at the voltage limit to obtain maximum power from the solar panel, which is what you want for MPP.

    Thanks for pointing out what you found about bq24715, I will give this feedback back to the team.

    The bq24650 input quiescent current is dependent on the FET Qg. If the FET Qc_total is lower than 10nC, you will get less quiescent current, since you don't need big FETs for high power. I suggest you test on the EVM before deciding.

    Regards,
    Steven
  • 0 in reply to Steven Bangdiwala
    Prodigy 50 points

    Hi Steven - Thanks again for your feedback

    Regarding bq24650, since the I(AC)=25mA states the value is for Qg=10nC and the switching frequency is 600kHz, is it fair to estimate the switching power for the external FET at 600kHz x 10nC = 6mA ? In other words, even if we were able to use a very small MOSFET, the floor would be I(AC)=25mA-6mA = 19mA ?

    Also, my question is what will happen at light load, according to the datasheet this charger will go into DCM mode. Does this mean the switching is substantially reduced, so less I(AC) power consumed by the charger ? Is there any way to get at least an estimate about what we might see on I(AC) when the solar current is small, in the range 10mA - 25mA ?

    For bq25703, the datasheet explicitly provides the no-load I(AC) in the 2mA-3mA range (I assume this includes the power to switch Qg=4nC), this seems very impressive!

    Thanks very much - Gil

  • 0 in reply to Gil Winograd
    TI__Expert 8690 points
    Gil,

    You are correct, because bq24650 uses DCM for light load instead of PFM, then it will consume more current than the bq25703A.

    The bq25703A has very good light load efficiency, as you have seen from the datasheet quiescent current. When you use the bq25703A, be sure to remove the OOA functionality. This part may be what you are looking for if you have a very low power solar panel.

    Regards,
    Steven