Other Parts Discussed in Thread: , EV2400, BQSTUDIO
Hello,
We purchased some BQ35100EVM-795 evaluation modules and an EV2400 interface to evaluate the BQ35100 chip.
Since we need to measure currents up to ~50mA, with the highest resolution possible, we replaced the default Rsense (100mOhm) with a 2.2 Ohm resistor. The rest of the EVM is untouched.
When I try to go through the calibration process with the Battery Management Studio software, using a 100 Ohm load for current calibration and no load at all for all the other calibration steps (Voltage, Temperature, CC offset, and board offset), I face the following issues:
1) when I enter the value -36 mA (3.6 Vbat, 100 Ohm Rsense) in the Applied Current field and I try to start the current calibration, the process fails with the following error message: "Calibration - Calibration current entered is out of range. This value is in milliamps. A typical value is -1000 for discharge current". Given the value of our Rsense I cannot use 1A, though
2) however, if I enter the positive value 36 mA in the Applied Current field, I seem to work around this limitation and the current calibration process completes successfully. I can then see in the Data Memory > Calibration window values for CC Gain and CC Delta that are quite close to our Rsense value: CC Gain = 2073.043 mOhm, CC Delta = 2107.307 mOhm. When I try to write these values to the data flash, I get another error message: "Value is beyond minimum value defined for parameter.:Calibration.Data.CC Gain". However, if I do a power cycle, these values can be read back from the data flash, so I assume that the write operation was successful (or that the calibration process itself took care of storing the correct values to the data flash)
3) after calibration has been completed, I start the gas gauge in ACCUMULATOR mode with the GAUGE_START command from the Battery Management Studio software (GE pin is always asserted / jumped to the pull-up resistor) and let it run for a while in three different conditions, all of them with a fixed purely resistive load: a) 50kOhm, b) 25kOhm, c) no load at all (PACK- pin disconnected). In all cases, the BQ35100 seems to accumulate an excess charge, equal to an excess current of ~200uA. Please find attached three log files taken with the Battery Management Studio software:
- 202002120834_battery+none.log is for the no load scenario. The BQ35100 sees a current of ~200uA, I would expect no current
- 202002111434_battery+25kOhm.log is for the 25 kOhm load scenario. The BQ35100 sees a current of ~370uA, I would expect a value of ~145uA, so I still have an excess current >200uA
- 202002111135_battery+50kOhm.log is for the 50 kOhm load scenario. The BQ35100 sees a current of ~290uA, I would expect a valut of ~72uA, so I still have an excess current >200uA
I have the following questions:
4) could the errors I got during calibration have affected the calibration process itself? What would be the correct process in my case (as I said, the *only* difference from the nominal EVM scenario is the Rsense=2.2Ohm)
5) if the calibration is correct, what does the excess current/charge come from?
Many thanks in advance and best regards,
