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.

BQ25895EVM-664: BQ25895 as MPP tracking and USB input

Part Number: BQ25895EVM-664
Other Parts Discussed in Thread: BQ25895

We purchased the MSP-EXP430FR4133 and connected as instructed in the guide to our BQ25895 demo board.

Our product incorporates ~13W of solar (14V OCV) and 3x 3.7V LiON batteries in parallel. 

After watching the training video I wondered if you would suggest a method of charging with the solar panels when the solar is in motion or if the battery is being used heavily by the system? I understand that the algorithm uses the voltage in and the charge current to constantly improve the power and that it works best when stationary. I'm thinking that if our system is being used and its getting fluctuating energy from the solar, it may give false info to the algorithm and perform badly. Would you recommend that we switch from the P&O method to F.OCV method when the system is in non ideal conditions? We could monitor our motion sensor as the trigger to switch modes I'm thinking.

Second question. If we still want to use the battery charger with a standard USB charger, would you recommend we use an external Mux controlled by the CPU to switch between solar/USB sources? Any other alternatives to using a mux?

Best, John

  • Hey John,

    The algorithm described in the training video works fine during movement and solar energy fluctuation, within reason. If the fluctuation occurs in less than something like 5s, then yes, it could falsely select a MPP until the next sampling cycle. The other note about this algorithm is it's limitations. When the charger is either in Constant Voltage mode of charging, the current into the battery cannot be accurately controlled. Additionally, significant loading can also indicate a false MPP.

    The critical note about this algorithm is that, unlike typical MPPTs which maximize input power, this algorithm focuses on maximizing charge current when in the Constant Current mode of charging.

    Now to your question of non-ideal conditions, I would suggest two options. The F.OCV method should provide a rough MPP as conditions change. Sampling time will be significantly shorter as you will only have to put the charger in High-Impedance, read the ADC, and exit momentarily (likely less than 500ms). Alternatively, for better accuracy but slower tracking time, you may consider writing an algorithm using the Input Current Optimization (ICO) feature of the bq25895. This would be on par with a true P&O MPPT, calculating the maximum input power. We have not yet developed the algorithm for this, but it is a possible method for these non-ideal conditions.

    In regards to a power mux, that is a good solution to use in order to switch between sources. Alternatively, you may use two PFETs to achieve something similar, but more control logic/software would be needed to prevent shoot-through, select a source, and correctly drive the FETs.


    Regards,
    Joel H