BQ78350-R1: Measured temperature to low

Martin Marmsoler1

Part Number: BQ78350-R1
Other Parts Discussed in Thread: BQ78350

Hi,

when checking the measured fet and cell temperature and checking with a thermal camera, we see at about 60°C about 5-10°C error. I've seen that our temperature model has still the default values, so I used your Thermistor Coef Calculator to recalculate the coefficients.

What is Adc0? The value at 25°C? So when I have a 10k NTC I have at 25°C an ADC voltage of about 1.65V --> ADC value = 2^14/3.3 * 1.65 = 8192?

When I change the coefficient (even to unrealistic values) the temperature do not change. Do I have to set anything else?

4657.BASE_CONF.zip

over 4 years ago

0 Matt Sunna over 4 years ago

TI__Guru* 81035 points

Hi Martin,

ADC0 should not be changed from the default value. It is based on internal resistance of the IC. Do you have the device configured to use the external thermistors or are you by any chance reading the internal IC temperature value?

Regards,

Matt

0 Martin Marmsoler1 over 4 years ago

Prodigy 175 points

Hi Matt, ok I let ADC0 as it is. Yes T1 and T2 are configured as external thermistors. I added the srec file above

0 Matt Sunna over 4 years ago in reply to Martin Marmsoler1

TI__Guru* 81035 points

Hi Martin,

Can you send me your data flash settings (.gg.csv file)? I'm not sure why you would not observe a change when playing with the temperature coefficients, so it would be good to check if there is anything that could explain this.

thanks,

Matt

0 Martin Marmsoler1 over 4 years ago

Prodigy 175 points

I added a screenshot. I export a gg file as soon as possible. I uploaded a screenshot. Cell Temperature model is the old, Fet temperature model are the new parameters

+1 Matt Sunna over 4 years ago in reply to Martin Marmsoler1

TI__Guru* 81035 points

Hi Martin,

Can you try setting the [SOURCE] bit in the Temperature Enable register to '1' to see if this makes a difference?

0 Martin Marmsoler1 over 4 years ago

Prodigy 175 points

Hi Matt,

yes this solves the problem. We disabled that bit, because otherwise there was high noise on the measurements. I will check out again and open another issue for the voltage ripple

But when SOURCE is 0, which internal temperatures are used?

Found the solution: The internal die temperatures of the BQ769x are used

When I set the parameters from our temperature sensor, the temperature is shown wrong.

In the following paper you use 18000Ohms for R_TH, but the internal pullup in the BQ769x is 10kOhms. Why are you using 18000Ohms in the spreadsheet?

Found the solution: No, 18kOhm must be used because of the BQ78350 firmware: e2e.ti.com/.../bq78350-thermistor-calculator-r

www.ti.com/.../slua924.pdf