BQ25619E: BQ25619e and shipping mode

Thank you for the clarification, anyway the BQ enters in ship mode even if the current is below this value, i set REG07 to 0x00 and i never modify it in any part of the firmware, it's very strange. I also monitor the battery voltage and it's about 3.9V. What could be the reason the BQ enters in ship mode?

Thank you

Hi,

What are your test conditions such as VBUS (voltage at VBUS pin), IVBUS, VBAT (voltage at BAT pin), IBAT, VSYS, ISYS right before the BATFET turns off?

Thanks,

Ning.

Hi, 

the power cable is not connected, VBUS=0Volt, IVBUS=0A, VBAT=3.8-4.0V (average value, we tested several values in the range) VSYS=VBAT, on ISYS we have several value, during normal runtime is 60mA, but when the GSM connection is active we have 1.6A, burst of 577us every 4.615ms.

Maybe there is some problem with current pulses?

Thank you so much for your support

Hi,

Did you monitor the BAT pin voltage during the current pulses? What is the lowest BAT pin voltage?

Thanks,

Ning.

Hi,

yes, i have connected the oscilloscope with 3.8V trigger, Vbatt never gone below 3.8V, even during the current pulses. Anyway, at random time, the BQ25 enters in shipping mode, i have 20 devices in test, and i have noticed some device enters in shipping more frequently than others. I also have tried a firmware that doesn't change the BQ25 registers, so it's in default mode of operation, so i don't think it's a firmware problem

Thank you 

Hi,

What are all the register readings before and after the device enters shipping mode respectively?

Thanks

Ning.

Hi,

i also have tested a firmware that doesn't change anything in the default settings, so the registers' values are the same indicated in the datasheeet, except for REG07, where the shipping mode is set to 1. Reading the registers after the problem is quite difficult since the microcontroller doesn't work since the bq25 cuts off the power supply in shipping mode. 

Another information, we have noticed that the problem arises more often if the GSM signal is low. In that case the GSM module requires more current. It's possible that the BQ25 is not fast enough to follow the current pulses? The maximum current pulses of the modem is 1.6A, so i don't think it's a problem of the absolut value, maybe there is a problem with the frequency? We have previous versions of the same device using BQ24195 and BQ24298 and we didn't have this kind of problem (thousands of devices on the field, so the circuit is validated)

Thank you

Hi,

Could you please provide the step-by-step test procedure to trigger BATFET turn-off?

Thanks,

Ning.

Hi,

the BQ25 output is connected to a GSM modem and to a microcontroller. The microcontroller is connected with the modem with UART and with BQ25 with I2C. At runtime, usually when the modem communicates with the cloud, the BQ25 goes in shipping mode without any reason, even if the firmware doesn't write anything on the I2C bus. The modem has 1.6A pulses. It's impossible to preview when the BQ25 goes in shipping mode, sometimes it works for hours, sometimes it goes in shipping after few minutes.

We also performed a second test. We have detached the Modem and the micro from the board, we connected two capacitor on the SYS lines, pratically nothing is connected on the BQ25, except the components required by the datasheet. No power input, no battery connected. As soon as we connect the battery on Vbatt, the BQ25 enters in shipping mode, you can see the measure of Vsys at the boot, it goes to 2V and then goes down, and when the capacitors discharge the voltage goes to 0 because the BQ25 is in shipping mode. In this second test we are pretty sure we are using default settings and there are no problems with modem or microcontroller. You can see the schematic of the circuit in the image

Hi,

For "No power input, no battery connected. As soon as we connect the battery on Vbatt, the BQ25 enters in shipping mode" condition, could you please take waveforms on both VBAT (the BAT pin voltage) and IBAT (battery discharge current) in additional to VSYS waveform?

Thanks,

Ning.