Other Parts Discussed in Thread: BQSTUDIO
Hi everyone, I've been reading a lot about the bq34110 and I had many questions for the calibration part. We now have a plan for the calibration, and I would like it if you could tell me if our method is valid.
First off, our system will be run on 2 AA alkaline batteries and will not consume too much current (max current is lower than 1A). I've read on a thread that the board offset is useful to reduce the CC errors at high currents. Since we won't have high currents, is this calibration necessary?
Anyways, here's our plan. Note that we will not use bqStudio, but rather an I2C interface with an MSP430 for calibration and flash programming.
A) Board Offset calibration. We skip the CC offset, as the datasheet tells us that calculating board offset also calculates CC offset.
So, our plan would be one of these methods :
1. Short Rsense and perform the Board Offset calibration routine
2. Assume that our PCB is consuming less than the minimum readable current and calibrate at zero
3. Put a power supply at 3V between VBAT+ and GND (instead of VBAT-), bypassing Rsense
B) Temperature calibration. We will have a Semitec 103AT thermistor on the board. Note that our product will be used at room temperature, so I don't know how critical this test is
1. Read the temperature of the external thermistor and the internal temperature sensor and do an average of both to calibrate
C) Voltage calibration. Since our maximum VBAT will be 3V, we won't use an external voltage dividor, so we will keep the default values for the voltage divider register in DF.
1. Apply 3V with a power supply between VBAT+ and VBAT-, compare to the value read by the bq34110
D) Current calibration. Note that our load will be less than 1A, since we don't have this kind of current on the PCB. We have two possible options:
1. Enable certain devices and calculate the current that should be needed from the battery. Compare avgRawCurrent to this theoretical current and adjust ccGain and ccDelta
2. Enable certain devices and read the voltage drop on Rsense with a multimeter. Calibrate current with this value.
We will perform this calibration on ~10 PCBs and average the values to create a golden image. What do you think, are our methods valid for calibration?
Thank you,
Fred
