BQ24715: Battery is discharging when an adapter is re-plugged in

Hello Christian,

The QB1 and QB3 are N-channel MOSFETS, not P-Channel MOSFETS.

Is there any other suggestion?

Thanks,
Kind Regards

Hello Lumina,

I have a couple of questions:

Can you measure the switch node waveform when you plug the in the adapter with 95% charged battery?

Is there a system load attached, If so what is the system load?

Best Regards,

Christian.

Hello Christian,

Pls refer to the below for the waveform of the input signal:

1) Normal charging waveform.

.

2) Abnormal (battery in discharging state) charging waveform.

3) Abnormal charging waveform when the adapter stars to power supply after dely 10s by pluged in the adapter.

The system load is the operating power of the CPU at full power load, and the rated power of the CPU is only 6W.

By the way, we also tried to add two 270uF/16V electrolytic capacitors to the VSYS output, but didn't work either. Our EC engineer also confirmed that the EC did not modify the register of the BQ24715 when the adapter was re-plugged in.

Thanks,

Kind Regards

Hi Lumina,

It looks the battery voltage is near BATOVP. Do you have a TI EVM to check its performance?

https://www.ti.com/tool/BQ24715EVM-115

Regards,

Tiger

Hi Tiger,

We will contact the local TI FAE to apply the BQ24715EVM-115 for testing.

By the way, we have some questions about the BATOVP of BQ24715. 

Does it mean that the BQ24715 enters BATOVP mode when the SRN voltage is higher than 104% of the set maximum charging voltage? If yes, does the level of PIN11 (BATDRV) jump from low to high at this time? Then the battery will go into a discharged state and the level of PIN11 goes back to low?

Thanks,

Kind Regards

Hi Tiger,

We also measured the waveforms of BATDRV during normal charging (adapter power supply for the battery and the system load) and abnormal charging (adapter and battery power supply for the load) when the adapter is plugged in momentarily, and found that the voltage on VBATA(SRN) will be up to more than 104% of 8.7V (the max. charging voltage), the voltage on the BATDRV pin will be pulled up. The waveforms are shown below:

1) Abnormal charging waveform

2) Normal charging waveform

From the waveform we can see that there will be a level change for the BATFET pin when the adapter re-inserted. But there will be two different charging states, the normal power supply of the adapter and the mixed power supply of the adapter and battery.

Thanks,

Kind Regards

Hi Lumina,

That is right, the BATFET will turn-off when BATOVP triggers. In this case, the BATFET gate voltage will rise since this is a PFET. 

Furthermore, I did not see ACFET/RBFET in the schematic. This FET together with BATFET provide a power path feature. It selects the power source for the system.  

Regards,

Tiger

Hi Tiger,

There is an ACFET/RBFEF on the circuit, and we are using the EC to control the ACFET/RBFEF, nstead of using the ACDRV pin of BQ24715 to control the MOSFET.

Thanks,

Kind Regards

Hi Lumina,

Please also look into the VBAT OVP when charging is first initialed. See red circle below. I don't expect VBAT to have a voltage jump if battery is connected. 

Regards,

Tiger

Hi Tiger,

Do you have any suggestions on how to avoid the VBAT OVP when the adapter was pluged in?

In addition, we found a new problem in the testing. That is, when the notebook was in idle mode (boot into the system, null of the software was in working, just keep it in this status), connected a 12V/3A adapter for charging to power 100%. Keep the adapter connected (charging current setting is 2.4A, but the measured charging current is about 500mA), 30min late the EC will stop charging, continue to keep the adapter in and we can found that about four or five hours later the battery power will fall from 100% to 99%.

Theoretically, in idle mode, the overall power consumption of the notebook will be very low, and the adapter is perfectly adequate for the system to work, the battery should not be discharging, but now it is the battery fully charged will occasionally discharge, which is not normal. Could this phenomenon be related to the previously mentioned anomaly of mixing adapter and battery power?

Thanks,

Kind Regards

Hi Lumina,

For BATOVP, my suggestion is to replace the BATFET, use the same one as TI EVM. The current PFET has a very slow turn off. 

To avoid drain battery in idle mode, (1) you can terminate charging when the target is reached, or (2) intentionally raise the min system voltage (higher than battery voltage), in order to avoid turn-on the PFET which drains battery. 

Regards,

Tiger 

Hi Tiger,

We replaced the BATFET with better parameters and re-plugged the adapter in heavy load mode at 96% and 100% battery power respectively, and found that both still in mixed power mode,which means the adapter and the battery are power for the system. The waveform are shown as below:

1) Waveforms of re-plugged adapter at 96% battery power after replacing BATFET with better parameter.

2) Waveforms of re-plugged adapter at 100% battery power after replacing BATFET with better parameter.

As we can see from the waveform above, with the battery at 100%, unplugging the adapter and reconnecting it momentarily, the VBAT (voltage of the SRN) did not exceed 104% of the set output voltage (8.7V), and there was no trigger for the BAT OVP, but it still could not go into the normal power supply mode (the adapter's output current is about 0.4A).

Thanks,

Kind Regards

Let us circle back to one of my earlier question. 

Do you have a TI EVM to check its performance?

https://www.ti.com/tool/BQ24715EVM-115

Regards,

Tiger