INA229 to measure temperature

Question

I am basically using INA229 to measure current, voltage, power, and temperature of four 12V 100Ah batteries connected in a series-parallel combination. But before I ordered my sensors I read from the data sheet that it has an integrated temperature sensor, since it serves my purpose I didn't bother to read more. Apparently, when I read it later fully I got to know it can only sense/measure the die temperature of the sensor.

So my question is: Can the sensor be modified (CODE MODIFICATION) to measure/sense the temperature of the battery along with current and voltage? If not, Can a temperature sensor be interfaced to measure/sense temperature? or Should I replace the INA229 completely with another device that helps me measure all three current, voltage, and temperature? (If any please recommend me)

Thanks

Answer 1

If you only need to measure above a threshold temperature for saftey then something like that or measure within 10C then that could work. If you need to measure a specific temperature more accurate than -say- 10C I would say probably not, the INA229 will put out heat of it's own. In addition, you would have to have a good thermal design and place the INA229 right next to the battery, with the package touching the battery. I would opt for an NTC thermistor or a digital temperature IC. Those would need to be placed next to the battery, probably with insulation so the air temperature is not affecting the measurement too much. But this all depends on how much accuracy you need.

Answer 2

The following assumes you do not want to add more components to measure the temperature (such as an additional SPI temperature sensor).

If you provide a good thermal connection between the INA229 and your battery (maybe from the ground pin and/or the package to the battery case) and put the INA229 for some time into sleep mode to get rid of self heating and then immediately after wakeup take a temperature reading - this may be close enough to the actual battery temperature. Of course you wont be able to communicate with the INA229 during this cool down time. You may not need to measure the temperature too often as the battery temperature might not change very rapidly.

It is vital though to avoid any additional heat from the shunt resistor to reach the INA229 which needs some care in circuit layout. Take a look if drawing large current from the battery increases the INA229 temperature reading a lot compared to zero battery current. If it changes notably the idea of using this approach will fail or be very inaccurate.

Also the INA229 should be isolated against surrounding air by some thermal isolation so that it is more tightly coupled to the battery case than to anything else. If done well one could get better accuracy by preparing a lookup-table from temperatures read from the INA229 and from a decent temperature sensor for correction. If you make these measurements you can also check if putting the INA229 into shutdown for some time makes a difference at all.