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.

Estimate of Accuracy/Resolution of ADS1248

Other Parts Discussed in Thread: ADS1248

We want to use  B-type thermocouple to mesure temperature with ADS1248. How to estimate accuracy/resolution(using ADS1248 internal ref and temperature is 1800c)?

  • Hello Frank,

    The accuracy that you may achieve is dependent on mainly three factors:  Accuracy of the Cold Junction compensation Measurement, accuracy of the Type B thermocouple, and accuracy of the Analog to Digital converter performing the TC measurement. There are many factors in your system that will affect each of the 3 items above; the following is an estimate/approximation based on some assumptions about the system and typical specifications of the ADS1248 device.

    A quick estimate/example:

    1)  Accuracy of the Cold Junction Compensation.  Let's say that the thermocouple is connected through a thermocouple connector/isothermal block and the cold junction compensation temperature measurement is performed using class A PT-100 RTD. Class A RTDs can achieve accuracies of +/-0.15C.

    2)  Accuracy of the B-Type Thermocouple.   Lets assume that the Class B thermocouple guarantees +/-0.5C accuracy.  (This info provided by the thermocouple manufacturer).

    3)  Accuracy of the A to D system.  The DC accuracy will be primarily a function of the Offset error, Gain error and INL error, and accuracy of the voltage reference.

    For example:

    A Type B thermocouple @ 700C produces a voltage of 2.431mV at the inputs of the ADS1248.  The user sets up the ADS1248 with the extenal Reference REF5020=2.048V and with a PGA=128V/V.  We can do an estimate of errors at a few points in temperature.  The seebeck coefficient (uV/C) of the thermocouple changes as function of temperature:

    Errors due to the ADC (based on t typical specs):

    offset error = ~2uV (typ)

    Gain error = 0.005% => 2.431mV*0.005%= ~0.1215uV

    INL error = 6 ppm => 6 ppm * VREF/PGA = 0.096uV

    Reference REF5020 Initial Accuracy Error = 0.05% => 2.431mV*0.05%=~1.215uV

    Measurement Error of ADC = square root( INL Error2+Gain Error2+offset Error2 + Reference Error2)

    Measurement Error ADC = 2.345uV at 700C

    We know that the Seebeck coefficient of a B-type thermocouple at 700C is approximately  6.8uV/C; using this coefficient we can do a conversion to temperature: 

    Measurement Error ADC in Celsius = 2.3415uV/6.815uV = 0.344 C

    Finally the total error in the system:->

    Total Error @ 700C = square root( ADC Errror2 + CJC Error2 + TC Error2)

    Total Error @ 700C = ~+/-0.6C

    We can repeat the same estimate calculations using a spreadsheet for different temperatures to obtain an extimate, please see below.  At 1200C, the calculated error in this example is approximately +/-0.8C.   

     Thank you and Best regards,

     

    Luis