BQ25185EVM: Using battery as a charging source

Hi Sean,

Charging from one battery directly to another can be a bit challenging since there is an inherent voltage drop from the input of a charger to the output of a charge do to voltage drops across the internal FETs. Typically for this kind of system, the source battery's voltage should be boosted a bit higher in order to enable the secondary battery to be fully charged.

What charger are you using for the source battery? Does it have boost capabilities such as USB OTG? Would it be possible to include some boost converter in your system?

If not, we may be able to have solutions with lower dropouts (depending on your charge current), but it is not likely to fully charge one battery from another using only step-down DCDC converters.

Best Regards,

Juan Ospina

Hi Juan,

Thank you for the information.

The source battery's charger is a proprietary IC, which I prefer not to involve in the secondary battery charging circuit. Due to our compact system design, we aim to use a single 1-cell charger IC for charging the secondary battery. Therefore, adding an additional boost converter is not ideal.

Could using a higher specification for the source battery (3.8-4.35V) and the secondary battery (3.7-4.2V) be a viable workaround to address the lower dropout and sleep voltage thresholds?

I am also considering the BQ25155 charger IC, as it appears to have low sleep voltage thresholds. However, I am uncertain if its dropout range is acceptable for our application.

I would appreciate any input on our current approach and the suitability of the BQ25155 charger.

Best regards,

Sean

Hello Sean,

We are looking into your additional information and will provide feedback later today or early tomorrow.

Sincerely,

Wyatt Keller

Hi Sean,

Sean Yeo said:

Could using a higher specification for the source battery (3.8-4.35V) and the secondary battery (3.7-4.2V) be a viable workaround to address the lower dropout and sleep voltage thresholds?

This could help to reduce alleviate some of the concerns, so long as the source battery is well charged (> 4.3V) , but ultimately the ability to charge the secondary battery up to full voltage will be hampered by the source battery voltage. So even if your source battery capacity is much larger, your secondary battery may not receive a full charge without some voltage boosting.

The BQ25155 can also work for this application though it might not entirely resolve the issue with sleep, the BQ25185 has a sleep voltage of 88 mV so I'm not sure if the 300mV limit your are seeing is necessarily due to just sleep. Can you provide the charge current for your application and if possible a schematic of your charger? I'd like to check the behavior you are seeing on bench.

Best Regards.

Hi Juan,

Thank you for the insight on voltage boosting.

I don't have a schematic aside from the one provided by TI, as I have been relying on the EVM for testing so far.

Here are my observations with the BQ25185EVM, where I used a power supply at VIN (to simulate the battery source) and connected the secondary battery (3.897V) at VBAT. The setup includes a 2K ohm resistor at Riset and a 24K ohm resistor at Ilim/Vset:

• When VIN is 4.2V, charging occurs with 0.013A drawn (at supply).
• When VIN is 4.19V, charging occurs with 0.001A drawn (at supply).
• When VIN is below 4.18V, charging does not occur, even when the /CE pin is toggled.

I agree that the issue might not be solely due to the sleep mode. Based on the datasheet, the condition for charging to occur depends more on VIN relative to Vbat+Vsleep and Vbat+Vindpm. It states that Vindpm needs to be 300mV higher than Vbat. While I am unsure whether INDPM can be disabled, I am considering Vsleepz (max exit sleep threshold) of 262mV as the next condition to enable charging.

Could you please confirm if my approach is correct? Additionally, I would appreciate your inputs on this issue.

Regarding the BQ25155, it does not seem to have the same conditions as the BQ25185. The datasheet states, "When a valid input source is connected (VIN > VUVLO and VBAT+VSLP < VIN < VOVP), the state of the CE pin determines whether a charge cycle is initiated." Its exit sleep threshold is also much lower at 130mV.

Regards,
Sean