BQ24610EVM: Charge Voltage Set is not pulling up correctly.

An update on this situation. I remeasured the impedance from BATT+ to PGND and it was not OL it was several hundred kOhms based on the impedance network of the board, but I don't think this is an issue.

I turned on the TS 1.8V CV supply. I turned on power to the board at 24V CV. PG LED turned on. I enabled CHGEN, CEN LED turned on. I measured my impedance from BATT+ to PGND and it was 12V. I waited 15 minutes. The voltage had come up to about 15.5V. 

Another update. I tried the same procedure turning on the device, then measured the voltage across the battery output (no battery connected again). The voltage I got was above 17V, but was very volatile, sometimes jumping down to near 15.5V and not maintaining a constant voltage. 

Hi,

First thank you for your time looking into my question. When following the procedure 3.4.2 for the charger regulation voltage listed in the user guide manual for the BQ24610 EVM module PS#1 is set to 24V, PS#2 is set to 1.8 V, Load 1 is off, and Load 2 is off. In this situation there would be an open load condition across the battery terminals which would represent a absent battery detection condition, yet the indicated expected voltage is still 18V +/ 1V for the Batt+ terminal with respect to PGND on the device. My confusion is that I am getting a more variable voltage at this pin with respect to PGND. When I first turned on the system I saw 12 V. I turned the system off, removed and replaced the resistors back on and got 6 V. Still having an issue, I removed C22 and got 12 V. I left the charger running in battery absent condition for about 20 minutes and the voltage slowly rose up from 12 V to 15.5 V. I turned off the power supply. The next day I turned the system back on and saw 17V on the power supply probably sporadically adjusting the voltage of about +/- 2V. Now this voltage supply might be high enough to charge my battery, and I suspect once the IC goes into CV mode it wouldn't be an issue, but I am worried about the state of the device because it doesn't seem that it is keeping its expected voltage. I have used another EVM module in the past which seems to work as I would expect, and maintain 16.8V +/- 1V in battery absent condition very well. 

Hello,

Please send waveforms of the input voltage, PH node, and output voltage with no battery. Please observe the STAT1 and STAT2 pins.

Thanks,

Mike Emanuel

Hi,

Attached is the voltage characteristic of the input voltage, PH node, and output voltage with no battery. The voltage of the PH node and battery output voltage appeared to be matched. STAT1 and STAT2 LEDs appeared off throughout the test. CE LED turned on when I jumpered the 3.3V from Vref to the pin, then turned off when I took off the jumper. Vfb pin seemed to be steadily regulating the voltage at 2.1V. I can perform a longer test if needed.

Thank you for your help,

James

Dear James, 

STAT1 and STAT2 being off indicates "Charge suspend, timer fault, overvoltage, sleep mode, battery absent."

Also, I need to see the switching of the PH node and the output. The converter switches at 600 kHz; a time scale of roughly 5 second steps misses a lot of information by sampling error. Is there any way we can see a scope capture of the same waveforms where 2-3 switching periods are seen and 2-3 period of the output are seen (if they are different)?

Thanks,

Mike Emanuel

Hello,

I understand that STAT1 and STAT2 LEDs being off is the expected behavior of the board when the charge enable pin is set but there is no battery connected thus satisfying the battery absent condition. Please correct me if I am misunderstanding your statement. Attached is a waveform showing the PH node in purple and output (BATT+) in yellow. 

Thank you,

James

Please also see the graph shown below, which is happening frequently and repetitively lasting for a few milliseconds.

I have added some more graphs showing the voltage characteristic when a trigger occurs on a falling edge on VBATT-. Note,  red is the FFT of the PH node. 

Hello,

I will verify this behavior on the board later this week.

Thanks,

Mike Emanuel

Hello again,

Based on the graphs that you have duplicated, I am under the impression that you believe that my evaluation module is operating under normal conditions and that differences in voltage over time I am detecting in my multimeter are from sampling error. Further, the voltage drop that is seen on the PH node is caused by Vfb pin to detect that the battery is absent. Am I interpreting your responses correctly? 

Hi, 

Back to the characteristic over the 5505s test, I have duplicated the test set-up with a sampling rate of 6 uS over about 23 minutes. The total amount of data points collected during this time was approximately 230 million. I trimmed down the data to 230,000 points representing samples every 6 ms to allow reasonable speeds for memory transfer. I have verified the 6 uS and 6 mS outputs match (for the most part). While the PH node characteristic may match expected behavior, it seems that the maximum and minimum values are changing over time. The first time I turned the board on the maximum voltage was between 10-12 volts, then after ten minutes the max voltage was about 17 volts. I performed a series of repeated tests turning on and off the input power to the board, and noticed that the voltage started significantly higher but trending downward to about 17V. At this point I am not considering the minimum voltage because I understand that this is being caused by the Vfb pin during the battery absent condition. I am thinking the voltage behavior on the output is caused by increased temperature on the board.

Would you be able to verify this type of behavior? I am concerned the built-in over-voltage protection will not work properly if temperature is causing the impedance network (voltage divider) to change over time. I am also concerned that it takes 10 minutes to initially reach the set voltage (16.8V) based on the impedance network. If the charger detects an under voltage condition, then the battery will not begin charging. 

Thank you for your time and consideration,

James