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.

Issue measure temperature TI Design Analog Engineer's Circuit: Amplifiers SBOA323–December 2018.

Other Parts Discussed in Thread: LMV324

Dear Sir ,

I am measuring the temperature through NTC . I prefer the TI Design Analog Engineer's Circuit: Amplifiers SBOA323–December 2018.Temperature sensing with NTC circuit. I gone through the design in which measuring range select -20 to 105 Degree . Based on Selected NTC value we found the R1 7.54K . We selected minimum output is 0.5V & Maximum Output 2.7V .

But As per calculation we found at -20 Degree 0.237 & at 105 Degree Found 3.07V . Reference voltage i set thorough voltage divider from 3.3V . That is found 1.56V .

Calculate the gain found 0.78 . Then according R3 Found 220R with respect to R1 1K .Actual 1.22 .

As per this gain value I am not able to found the output as desired after gain select .

Please suggest . 

  • Former Member
    0 Former Member

    Hello Mohit,

    Do you have a schematic you could share?  I would like to confirm that I have understood you properly.

    Did you calculate the output range by hand, by simulation, or by physically building the circuit?

    One thing you need to take into account is that the temperature range is quite wide.  Thus, there may be some drift in the values of the non-NTC resistors as well.  Please send a schematic and we can go from there.

    Regards,

    Daniel Miller

  • Hi Danil ,

    Thanks for your reply.

    Yes I calculate the range on by hand by taking reference of  TI Document reference Amplifiers SBOA323.

    I am sharing the schematic with you  . Please check and revert . I am sharing the NTC temperature table also for your reference.

    Reference voltage I prepare by using simple voltage divider .

     

    In This Circuit R307 , R305 Consider 0 ohm .

    R306 is calculated by root of Rjmin X  Rj max by given below table .

    Temp (°C) Resistance in OHM
    -20 96607
    -15 72809
    -10 55253
    -5 42292
    0 32640
    5 25390
    10 19901
    15 15713
    20 12490
    25 10000
    30 8057
    35 6532
    40 5327
    45 4364
    50 3598
    55 2980
    60 2484
    65 2080
    70 1750
    75 1476
    80 1253
    85 1070
    90 916.5
    95 787.7
    100 679.7
    105 588.3

    Looking forward from your side . 

    Regards

    Mohit 

  • Former Member
    0 Former Member in reply to MOHIT SHUKLA

    Hello Mohit,

    I've reviewed your schematic and have several comments for you.  Please note that I've assumed your NTC is on the top of the resistor divider.

    1.  In the high-voltage scenario, the NTC impedance drops to ~600Ohms, which causes your input common mode voltage to approach 3.11V.  This is beyond the input common mode range of the LMV324, which can only go to about 800mV away from the high supply.  So, if you choose to use this combination of NTC with 10kOhm resistor, you will need an amp with a wider input common mode range perhaps even a rail-to-rail input.

    2.  To achieve your desired output voltages of 500mV at low temperature and 2.7V at high temperature, you will need two different feedback impedances.  You cannot use the same feedback resistor and get the desired output voltages. I suggest you follow the process described in the cookbook (SBOA323).

    3.  Having not run a stability simulation, I can almost guarantee that the circuit will not be stable without some form of compensation because the load capacitor of 0.1uF is quite large.  We discuss this issue and its solution in quite a bit of depth in TI Precision Labs.

    Please let me know if you have any further questions.

    Regards,

    Daniel

  • Hi Danil ,

    Thanks for your mail . please discuss and share the solution .

    Regads

    Mohit

  • Hi Mohit,

    your schematic is totally confusing. Can you please post a schematic which exactly shows (in every detail !) what you are doing, without the need of guessing and assuming?

    Kai

  • Former Member
    0 Former Member in reply to MOHIT SHUKLA

    Hello Mohit,

    Has this issue been resolved?

    I see you've marked my previous post as resolved, but I'm not sure that the issue is closed.

    If I don't hear back from you, I'll assume that my previous reply was sufficient.

    Regards,

    Daniel