BQ25756E: Suggest PMIC for 24V solar input and 24V Li-ion battery charging with I2C battery gauge + 5V/3.3V buck outputs

Hi Dhruvit,

• Input Source: The BQ25756E can charge at those ranges with MPPT
• Battery: The BQ25756E can charge a 24V Li-ion battery
• Charging: MPPT for solar panels is integrated into the BQ25756E
• Monitoring: The BQ25756E does not have monitoring.
• Output rails: the BQ25756E only has the one charging rail. the 5V REGN is for the gate driver and thermistor and has a current limit around 100 mA
• Power management: The BQ25756E does not have power path.
• Protection: The BQ25756E has over voltage, undervoltage, and over current protections for both the input and battery.

The BQ25756E is the best charger for the design requirements. TI does not have a single IC that can perform all of the system requirements. I think you will need a separate battery gauge and a separate PMIC to handle the 5V and 3.3V rails. Let me know if you would like me to loop in the experts on the battery gauges for a requirement in that space.

Best regards,
Michael Bradbourne

Hi Michael,

Thanks for the clarification. I have one follow-up question.

My solar panel outputs around 24 V (18–28 V range), and my 24 V Li-ion pack varies from 20 V (discharged) to 29 V (fully charged). From the simplified schematic, it looks like the system rail (VSYS) follows whichever source is active.

Can the BQ25756E regulate or fix the system voltage (e.g., 24 V) independent of battery or solar voltage — for example:

• When the battery is at 22 V (night, no solar), can the PMIC boost to maintain 24 V on VSYS?

• When the battery or solar is at 29 V, can it buck down to 24 V?

• Can this regulation be set or controlled via I²C commands?

If not, could you please confirm whether any other TI PMIC supports this kind of bidirectional buck-boost operation with configurable system voltage and solar MPPT charging?

Thanks,
Dhruvit

Hi Dhruvit,

Based on the image that I see here, the system is directly attached to the battery. This means that the system voltage will follow the battery voltage. If you would like a separate regulated system voltage, you will need another converter to regulate that voltage.

In general, the BQ25756E can only regulate input current, input voltage, charge current, or charge voltage. Only 1 loop can be active at any given time. While using MPPT, the device will regulate the input voltage, but this only works best when the only device connected to the solar panel is the BQ25756E. This means that the system would be in parallel with the battery and will follow the battery voltage. To have a regulated system voltage, it is best to have a converter stage between the battery and the system.

Best regards,
Michael Bradbourne

Hi Dhruvit,

I forgot to answer this part of the question:

Dhruvit Bhimani said:

If not, could you please confirm whether any other TI PMIC supports this kind of bidirectional buck-boost operation with configurable system voltage and solar MPPT charging?

There is not a battery charger that does two voltage rails and MPPT. MPPT requires 1 device per solar panel, which typically means 1 power output. The best way to do this would be to have a secondary power converter from the battery to the system. While charging, the secondary converter will take the input power away from the charge current, and the leftover current will naturally charge the battery.

Best regards,
Michael

Hi Michael,

Thanks — that matches my understanding. Quick confirmation request:

I couldn’t find an integrated FET PMIC that directly produces a fixed 24 V @ 8 A (≈200 W) from a 20–50 V range, so my current plan is to pair the BQ25756E (MPPT charger → VSYS) with a bidirectional buck-boost controller. I found the BQ25758A (I²C-controlled, up to 60 V, bidirectional controller with USB-PD features) and intend to feed VSYS from the BQ25756E into the BQ25758A input and have the BQ25758A regulate a fixed 24 V output for the system.

Can you confirm whether that is a reasonable approach? Any important caveats you’d call out ? If TI does not recommend BQ25758A here, could you suggest the preferred controller or topology for a ~200 W regulated VSYS?

Thanks again,
Dhruvit

Hi Dhruvit,

That should work. The BQ25756E will take the solar panel input with MPPT. The output of the BQ25756E will supply the power to the BQ25758A for the system load and the rest of the output power will charge the battery.

I would point out that the two devices have the same I2C address, and for this application, I2C is needed for both devices. It may be better to choose a different external FET DC/DC to save yourself the I2C mux or pins on the host controller, but the architecture makes sense to me.

Best regards,
Michael Bradbourne

Hi Michael,

Thanks — I’ll keep both devices but place them on separate I²C channels (or use an I²C switch) to avoid the address conflict. We'll use the BQ25756E for MPPT/charging and the BQ25758A as the 24 V regulator only (no PD negotiation), with configuration at startup.

Before I finalize the schematic, any concerns about startup sequencing, I²C pulls/leveling, or control-loop interactions between the charger and the bidirectional controller I should watch for?

To conclude this thread, thank you for your help and detailed guidance throughout. It really helped finalize my system architecture.

Thanks again,
Dhruvit

Hi Michael,

While going through the datasheet, I found a comparison table of the BQ25 series. It shows that the BQ25750 and BQ25820 devices support direct power-path management. Among them, the BQ25750 also offers a buck-boost topology, which seems ideal for my use case where the system requires a constant 24 V output (VSYS) even when the battery voltage is lower.

Can you please confirm my understanding below:

1. When no input source is present, the BQ25750 will discharge the battery and regulate VSYS to the programmed 24 V output.

2. When an input (e.g., solar panel) is present and its voltage exceeds 24 V, the device will still regulate VSYS to 24 V and charge the battery in parallel — rather than passing the higher input voltage directly to VSYS.

3. The VSYS rail can deliver sufficient output current (around 8 A) as long as the external components are rated accordingly.

If the above understanding is correct, then the BQ25750 alone should meet my system requirement, and I can eliminate the BQ25756 and BQ25758A from the design. Please confirm before I proceed with ordering the development board.

Best regards,
Dhruvit