Thermocouple Calibration

            A thermocouple can be any junction between two different metals and may be used to measure temperature. Each metal produces a different electrical potential that varies according to changes in temperature. This rate of change is different for each of the metals in the thermocouple, so a thermocouple produces a voltage that increases with temperature. You can calibrate a thermocouple by plotting the thermocouple's voltage-temperature curve.

Apparatus: 
Thermocouple; 
Temperature controllable bath (thermo bath)
Digital Thermometer;
Multimeter.

Procedure: 

Step 1
         Fill the thermo bath container with water and turn the thermo bath on. Heat the water to 30 degrees Celsius and turn the thermocouple device on. Connect each lead of the multimeter to one end of the thermocouple. This multimeter should be able to measure a voltage of 1 microvolt.

Step 2
         Place one junction of the thermocouple into the water and allow the voltage to stabilize. This occurs when the voltage stops fluctuating except for the last digit. Record the stable portion of the voltage from the multimeter.

Step 3
         Increase the water temperature to 35 degrees Celsius and record the stable voltage on the multimeter again. Repeat this procedure for each 5-degree increase in temperature from 35 to 60 degrees Celsius.

Step 4
         Measure the room temperature and look up the voltage for your thermocouple type at the room's temperature. For example, the voltage for a type K thermocouple at a temperature of 25 degrees Celsius is 1 millivolt. Add this value to each of the voltages you recorded in Steps 2 and 3.

Step 5
         Use the curve-fitting method of your choice to find the line that best fits your recorded data. The slope of this line provides the voltage increase for each degree of temperature increase. The voltage on a standard type K thermocouple should increase about 40 microvolts for every degree Celsius increase in temperature.

(As mentioned in the theory, the multimeter reading corresponds to the difference in temperature between the surroundings (room) and the bath. To calibrate the thermocouple, we have to take the room temperature into consideration to get the absolute value of temperature measured. Find the equivalent millivolt value for the room temperature from the corresponding Thermocouple table (E_RM). Then, add that millivolt value, corresponding to the room temperature, to every multimeter reading (E_MM). Tabulate the values.)

Observations (Type-K thermocouple used):
Room temperature : 20.5° C = 66.2° F; corresponding mV reading from tables E_RM = 0.818 mV

Bath Temperature,TB [° C]	Bath Temperature, TB [° F]	MM reading, EMM [mV]	Total emf, E = EMM+ ERM [mV]	Temperature corresponding to E, TTC [° C]	Temperature corresp. to E, TTC [° F]
30		0.361	1.179	29.4
35		0.579	1.397	34.8
40		0.788	1.606	39.9
45		0.995	1.813	44.9
50		1.202	2.020	49.9
55		1.407	2.225	54.9
60		1.613	2.431	59.8

Plot the measured bath temperatures values (T_B) on y-axis against the corresponding thermocouple emf (millivolt) values (E) on x-axis. Find the slope, intercept and the correlation coefficient of the curve-fitted line by any method. If the correlation coefficient is not very close to one, curve fit with higher order polynomial. Calculate the 95% uncertainty in temperature measurement of the calibrated thermocouple sensor if the obtained curve-fit correlation is used.

NOTE: 
The multimeter voltage reading should be zero when you measure room temperate without the reference junction (the thermocouple connected directly to the meter) Then the reference temperature is the same as measured temperature, both equal to the room temperature, therefore the temperature difference and generated voltage should be zeros. However, your reading will never be exactly zero due to many reasons:

1.     if the tip of the thermocouple is wet it may be sub-cooled due to evaporation;

2.     the room temperature is non-uniform;

3.     the meter is not perfect (make a short-circuit to check its zero),

4.     the measured circuitry may be picking up some "noise," since it acts as an antenna; etc.