BQ25798: The BQ25798 Vbus and charge current looks strange

Hi Nan,

Is there a load on SYS?  What does the status and fault registers in REG0x1B-0x27 report?

Regards,

Jeff

Hi Jeff,

There is small load on the SYS, - a STM32 MCU and two memory chips. I don't believe the load will draw that much power as the VBus is 12V,

so the power divided should be more than 10W, if this is really happened, then the MCU and other chips could be really hot, but they were not.

There is zero fault flags reported:

Hi Nan,

The ADC must be wrong.  Unless you have disabled IINDPM limit in REG0x14, IINDPM is 2850mA and you are measuring 3060mA, which should have charger in IINDPM.  But there is no IINDPM status. Are you measuring the input current with a meter?  Is it really 3A?

Is this on an EVM or your own board? If your own board, did you follow the recommend layout, esp the PMID and SYS 0.1uFcapacitors?

Regards,

Jeff

I do have the IIDPM enabled:

This is my own board, the PMID and SYS caps are showing below:

You think there are more 0.1uF caps needed?

Hi Jeff,

There was a mistake on the current reading, the ADC value is 1mA /bit but not 10mA/bit, I was misled by the configuration where the bit step is 10mA.

Sorry about that. 

So I measured the Ibus current on my board, it matched the ADC value red from register, thus, the ADC readings and capacitors have no problems.

But it brought out another issue, the charge current seems to be too low. The charge current is set to 2.8A(REG03->0x118) which I checked register with no problem or mistake, but the charge current is only about 147mA while it was in CC charge stage (the first 3bit of the Charger_Status_1: 011). 

The charger should charge the battery with the "Charge Current Limit" that I configured which is 2.8A in my case?

Yes, I did measured the charge current and IBUS current with multi-meter, to be sure, I just measure them again with the meter and got the following numbers:

IBUS-> 298mA(ADC=> 301) 

IBat -> 128mA (ADC->142)

Charge is in CC charge phase:

Here is all the register value that downloaded to the IC:

I have the Battery Management Studio register file if you need it I can send it to your email as this webpage won't allow to attach file.

Another strange thing is that, when I unplug the battery from charger, it went to the Pre-Charge phase which I think should be no-charge phase as the battery is unpresented. May I correct?

Hi Nan,

When battery is removed, charge goes to trickle charge ~=100mA into BAT.  Can you download the status and fault registers in REG0x1B through 0x27 as well?  They will report what is happening.

Regards,

Jeff

The Reg  0x1B to 0x27 is on the previous image I posted above. The Charger_Status_0 is 0x1B and 0x27 is on the fault_flag_1.

they are all zero

Update:

The issue is the /QON signal. It needs to be pull high to get the Vsys output, but by plug in the VBUS only (battery is not plugged yet), the Vsys

has no output, then there is no 3.3V can provide the pull -up power to /QON pin. 

In order to get it work, the battery needs to be plugged before the VBUS presents. Is this true?

In addition, I set the charge current to 2A, the chip was really hot, it will reach 100°C in 2 mins and trigged the thermal regulation , so I decreased the current down to 500mA, but still heated to 90°C. Is this normal?

Hi Nan,

/QON is internally pulled up.  If not used, leave floating.  To exit ship mode or force reset, pull low.  /QON is powered from battery if battery only or VBUS is VBUS applied.

Regarding thermals, what is the efficiency per datasheet efficiency curves?  The larger the VBUS to VBAT difference, the worse the efficiency.  The IC should be able to handle up to 3-3.5A without thermal issues.  How much copper area/pours/thick traces do the VBUS, PMID, SYS, BAT and GND pins connect to, even if via'd into internal or bottom layer?  More is better.  Changing to 750kHz and 2.2uH inductor will be more efficient.

Regards,

Jeff