BQ76952: TEMPERTURE reading error

Dear Jian,

Please correct me if I am wrong, but it seems like you have set up 5 thermistors on your device. This leaves me with a few questions that will help me understand the situation better:

1. Are all 5 thermistors seeing the wrong value at the same time or does one produce the proper value while the other 4 produce incorrect values?

2. Do the proper values ever reappear after going high or does it only go away after restarting the device.

3. Can you share your configuration of the temperature sense pins like the Settings:Configuration:TS1 Config register?

4. Could you share the hardware set-up of your thermistors? Is there any capacitors in parallel to the thermistors and what type of thermistor are you using?

Best,
Asher Nederveld

Hi  Nederveld,

    1.  following  is  reading  tempt

2.

3.  Software  config:

/* This configures the ALERT pin to drive high (REG1 voltage) when enabled.*/
/* The ALERT pin can be used as an interrupt to the MCU when a protection has triggered or new measurements are available*/
g_Afe_Config_ok |= BQ769x2_SetRegister(ALERT_PIN_CONFIG, 0x2AU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FDU, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FFU, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FBU, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x9301U, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x9302U, 0x0BU, 1U);

Dear Jian,

You're configuration for the pins looks like it should be fine. However, I wanted to double check that the thermistors are off the board of the schematic shown and mounted near the battery and that's why the capacitors are only showing to for the pin inputs.

When these changes in temperature occur, is there a significant different on the pin voltages? It would be helpful to know whether the corresponding change in temperature correlates to a change on the input pin voltage to determine whether the issue is on the hardware or software side.

Best,
Asher Nederveld

Hi   Nederveld,

       Because the  problem  is  diffcult  to   reappear,  I  want  to  do   something  for  this.

Dear Jian,

How often are these issues occurring? From the sound of your previous message, it seems like that it isn't occurring often. Once it occurs, does the temperature ever fluctuate or does it stay fixed at 96 degrees? If it doesn't fluctuate, that might point more towards a software issue with an incorrect read.

However, if it does fluctuate when at the high temperatures, that would make me believe it is more of a hardware issue. What sort of thermistors are you using in your set-up? If you are using BQStudio for you programming, you can send me the .gg file and I can try to recreate your issue on an EVM where the hardware is confirmed to work.

Best,
Asher Nederveld

温度传感器0_125度10334410 (10-G104-03).pdf

Hi   Nederveld,

     This  is  our  tempt

Dear Jian,

Thanks for sharing the thermistor with me. Nothing about it seems out of the ordinary. I am curious about the situation where the 96C temperatures occur. Could you clarify what exactly is happening? Is the temperature staying exactly at 96C or does it vary and change with time? For example, would applying external heat to the thermistor produce any change on the measurement when it is stuck at 96C?

I would recommend checking the voltage at each thermistor pin when these 96C temps are being recorded to see if an abnormal voltage is on the pin when compared to the normal readings after restarting the device.

Best,
AsherNederveld

HI  Nederveld,

        I  measure  the  voltage  between   ntc  sensor.  The  voltage   is  16mv.    Other  ntc  sensor  is  almost  10~20  mv.   Why  are they  so  low?  Today,  we   find   the  DFETOFF  is  16 degree,  other more than 30 degree.

Hi  Nederveld,

    I  think  maybe  the  ntc sensor  voltage is  too low . It  will  lead to  extra voltage  disturb the  sensor.  

Hi Jian,

If you're seeing ~10-20mV across the NTCs, this makes me think that the 180kohm internal pullup is being used for the pins.

Are you seeing this low voltage at all times? Or just during the fault?

Regards,

Max Verboncoeur

Hi   Max Verboncoeur,

  step1:

    When  NTC sensor  is  inserted, I  see  NTCs  voltage  low  at  all times.   But   When  NTC sensor   is not inserted, I  measure the  inserting point  voltage    and  get   950mv.    Would you  tell me  what is  the  normal  votage?

    step2:

   I check config, every   config is  0xB.    

     g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FDU, 0x0BU, 1U);
g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FFU, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FBU, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x9301U, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x9302U, 0x0BU, 1U);

This  is  our  customer  tell us  they  see  this.  The  temperature  is  room  tempt.  Sensor1~6  is  other sensor.   Sensor 7~11  is  AFE sensor. 

Hi Jian,

Just to confirm whether this is an issue with the ADC calculation or with the actual hardware, can you provide the raw ADC measurements from while the device is exhibiting this behavior?

You can find the registers to get these values from Section 4.6: Subcommands 0x0075–0x0077 DASTATUS5–7(), Additional Measurements of the TRM.

Regards,

Max Verboncoeur

Hi   Max Verboncoeur,

     The  test  is  going  on  and   does not  repeat  the   problem.  At  this  time ,  I  want  to  raise  ntc sensor  voltage.  Would you  tell  me  how to  config  the software ?

    Now   my  config  is  as  following:

    g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FDU, 0x0BU, 1U);
g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FFU, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x92FBU, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x9301U, 0x0BU, 1U);

g_Afe_Config_ok |= BQ769x2_SetRegister(0x9302U, 0x0BU, 1U);

Hi Jian,

See the image below for a diagram of how the pins function internally. You won't be able to raise the voltage from the pin since the 1.8V REG is the only pull up voltage available, and the 18kohm option is already selected.

Your software configuration should be fine, however, another possible test that you may try is using the example code provided on the product page to see if that has any effect on this behavior.

Regards,

Max Verboncoeur

Max Verboncoeur,

     I  find  something  strange 。     

I  find  when the extra voltage  is more than 120mv  we can read temperature data. When extra  votage is less than 120mv, we can not read valid temperature data.  But  when ntc  sensor  linked BQ76952, the  voltage  is  about 20mv.  It  is  so  strange

Hi Jian,

The TS pin works by periodically sourcing current on the pin, and simultaneously reading the voltage dropped across the external thermistor. Depending on how you are taking these measurements, the pulse may be too quick to read.

Can you send the raw ADC temperature readings? You can find the registers to get these values from Section 4.6: Subcommands 0x0075–0x0077 DASTATUS5–7(), Additional Measurements of the TRM.

Also, don't have anything extra connected to the NTCs (i.e. meters/probes) when taking these readings as too much capacitance on the pin could interfere with the measurement process (the 470pF capacitors should be fine to leave as they are).

You may also send your .gg config file so that we can check it.

Regards,

Max Verboncoeur

Hi Jian,

This is the thermistor config file. Could you send the raw ADC temperature readings from the device? You can find the registers to get these values from Section 4.6: Subcommands 0x0075–0x0077 DASTATUS5–7(), Additional Measurements of the TRM. I am looking for the TS1/2/3 Counts from DASTATUS6() and DASTATUS7().

Regards,

Max Verboncoeur