BQ76920: Invalid Current Fault

Hi Vinith,

What do you see when you read the SYS_STAT register while discharging with 2A? Which protections are triggered? And what are your settings for the PROTECT1 and PROTECT2 registers? Also, is this on your own board or on an EVM?

Thanks,

Matt

Hello sir,

I was reading my registers in decimal values and not hex values. SYS_STAT most of the time gave me SCD fault when I tried to draw 500mA with value of 130(128-CC_READY,2-SCD_FAULT). When I try to draw 2A it started giving me OCD fault with SYS_STAT being 129(128-CC_READY,1-OCD_FAULT). I am not using EVM board. I am attaching all the registers values I used and schematic of the board. I am attaching register values in the sequence of the program.

CELLBAL (0x01) - 0x00// I am using external cell balancing so I turned off the internal cell balancing

STS_CTRL1 (0X04) - 0X10//Only turning on ADC.

SYS_CTRL2 (0X05) - 0XC3//Turning on CC in one shot mode and turning on DSG_ON and CHG_ON bits.

PROTECT1 (0X06) - 0X9F//I set RSNS bit to 1 to double the range and maxed out on delay and threshold values of SCD.

PROTECT2 (0X07) - 0X7F//I kept the max values for the OCD delay and threshold values.

PROTECT3 (0X08) - 0X00//I didn't keep any delay for OV and UV faults.

OV_TRIP (0X09) - 0xFF// I set the max as OV fault.

UV_TRIP (0X0A) - 0X00// I set the min as UV fault.

CC_CFG (0X0B) - 0X19// Specified in datasheet.

SYS_STAT(0x00) - 0X7F//Clearing all the bits in the register.

This is the schematic used. Please help me on this. Any help would be appreciated. Thanks in advance.

Schematic_Battery_BQ76920.pdf

Hi Vinith,

I do not see an issue with the register settings or with the schematic. Can you measure the voltage across the sense resistor and the voltage as close as possible to the SRP/SRN pins while you have a 500mA load? Maybe there is high resistance on the board that is causing the current measurement to measure much higher. With 500mA, there should only be 2.5mV across the SRP/SRN pins. 

Best regards,

Matt

Hi Matt,

I tried to get the voltage drop across the sense resistor when load is connected, with CC enabled it is just going to fault as soon as I connect the load so there is no current flowing through the ckt. Hence I disabled the CC using 0x05 register and then measured the voltage across sense resistor I got 11.5mV for a load current of 2.28A which is exactly same. Can you please help me with this.

Hi Vinith,

It may be good to capture waveforms when the event is occurring. If there is any noise that is coupled onto the SRP and SRN pins, it can trigger false faults. Since enabling CC seems to cause the fault to occur, it seems there is a noise issue on the board. The SCD and OCD fault detection is done with hardware comparators, so the CC_EN bit should not impact this function.

The advice in this other thread may be helpful to debug this:

https://e2e.ti.com/support/power-management/f/196/t/873622

Best regards,

Matt

Hi sir,

I did the testings with the CC off sir I was able to draw more than 3.4A without any fault but as soon as I turn on the CC my upper limit came down to around 2.5A. Can I know what might be the issue sir. Because turning on CC can't shift the limit down right. It's not only the CC even the limits which I've set for the SCD and OCD are also pretty high sir, according to the datasheet. But it's triggering at a significantly lower current sir. Like I set my OCD current to 12A but it's giving fault at 1.7A itself sir. If I move the limit up to 20A it's giving me around 3.5A sir. How can this happen sir. I used all the resistances and capacitances according to the sample circuit specified sir. Can you tell what might be the issue.

Any help would be appreciated thanks in advance.

Hi Vinith,

This issue is really strange. The CC should not impact the current across the sense resistor. Have you tried replacing the IC in case there is something wrong with this specific device? 

Regards,

Matt

Hello sir,

I built the circuit in the schematic with half of it on the PCB and the rest of the part on the breadboard. Do you think that can cause this issues sir.

I even tried to replace the chip sir, I was getting around the same response sir so, I am suspecting the breadboard part of it might be accounting for the noise.

Vinith,

I didn't realize you were using a breadboard. This may definitely account for the inaccuracy you are seeing. The sense resistor needs to be very close to the IC with a good layout because any noise may trigger OCD or SCD.

Best regards,

Matt

Hi sir,
Can you please verify this layout with the schematic sir. don't worry about the routes sir. All the routes are perfectly fine and are kept in accordance with their current capabilities. please verify the positioning of the components sir.

Any help would be appreciated. Thanks in advance.Schematic_Battery_BQ76920_2021-01-27.pdfGerber_Battery_PCB.zip