This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

Connecting multiple thermistors in parallel

Other Parts Discussed in Thread: TMP107

Hello,

If i have 9 100K thermistors in parallel the effective resistance is 11.1K. Now the manufacturer datasheet of the 100K thermistor has a table describing the resistance at a specific temperature. If i use the same table the temperature values are incorrect, obviously because 100K is specified at 25 deg C. However if i use the 10K thermistor table from a 10K thermistor datasheet and use those values to compute the temperature of the 100K parallel branch, the values are correct. (The thermistor pull up is 10 K)

Just want to know if this is the right approach ?

This is a continuation from the topic below:

e2e.ti.com/.../686211

Thanks..

  • Revenant,

    Which thermistor manufacturer are you using? Part number?
    I don't understand how you are planning on measuring the temperature of 9 cells with the thermistors in parallel. Do you have a circuit diagram of your setup? Are you trying to measure the average temperature?

    Note: The manufacture data sheet gives you R(T)+tolerance. That is the individual thermistor model you need to build your customized equivalent Req(T). Ex. Req(T)=R1(T)|R2(T)

    -Kelvin
  • Hi Kelvin,

    Thanks for the reply.

    So, there are 9 thermistors on each cell in parallel to give the average temperature...

    The sensor is NTCG104EF104FTDSX - https://media.digikey.com/pdf/Data%20Sheets/TDK%20PDFs/NTCG104EF104FTDSX_Spec.pdf

    The pull up resistor  in the voltage divider configuration is 10K.

    Regards

  • Revenant,

    When you have 9 thermistors in parallel, the tolerances add up and this will impact your system accuracy. Have you consider using integrated temp sensors like TMP107? The system accuracy is guaranteed with digital sensors. You can daisy-chain these devices and you will only need 3 wires (vcc, gnd, and uart).

    -Kelvin
  • Hi Kelvin,

    No, i have not considered using the TMP107 yet but will definitely give it a shot. But i need to figure out the temperature computation aspects of this thermistor arrangement.

    My original question, however, was w.r.t computing the temperature from the above n/w.

    I am using the steinhart equation to compute the temperature after obtaining values from the adc.

    The steinhard co-eff i am getting from this link - www.thinksrs.com/.../ntccalculator.html

    - i use the values from datasheet to fill up this calculator. But when the computation is done the value at room temp is not as expected. If all resistors are at 100K at room temp ideally, the equation gives out a temperature of 78 degC. If you look into the table 78 degC is actually the temp when the resistance is 10K.

    is there any method to map into the 100K thermistor curve when using thermistors in parallel ?

    Regards
  • Revenant,

    You can built your parallel equivalent resistance model with the R(T) data:

    Example:

    T R1 R2 R_parallel
    0 1000 1000 500
    25 100 100 50
    100 50 50 25

    Now you have R_parallel as function of Temp. You can do this in excel.

    -Kelvin