ENGR 202 DUE IN 12 HOURS. HANDSHAKE REQUIRED

ENGR 202 Evaluation and Presentation of Experimental Data II – Summer 2016 Lab 4: Capturing Temperature Measurements with a

Thermocouple Original: Dr. Scoles, Dr Miller, Dr Chmielewski Rev: Dr. Marino

 

8/17/16 page 1 of 7

Goals

 Measure, plot, and record temperature measurements from a Type K thermocouple (TC)

 Correct the measured voltages with a calibration curve

 Find the time constants of the TC cooling curves Equipment/Software

 NI USB TC-01 Thermocouple Measurement Device

 Type K thermocouple, Omega KTSS-HH o Nickel-10% chromium (+) vs. Nickel-5% aluminum and silicon (-)

 Power resistor, 100 Ω, 25 Watt

 Hewlett Packard E3631A DC power supply

 Excel Reading or Viewing

• Review – Week 8 lecture notes Introduction

A thermocouple (TC) can be used to measure temperature over wide ranges in a variety of measurement environments and with fine spatial resolution. The sensing operation of the TC is based on the Seebeck effect: when two dissimilar metals are joined at both ends to form an open loop, an open circuit voltage is developed (Figure 1). The voltage is proportional to the difference in temperature at the two junctions. The measured voltage is on the order of tens of millivolts. To extract the temperature at the measuring junction (T1) from the measured voltage, we will want to keep the reference junction (T2) at a fixed, known temperature.

 

Figure 1. Two junctions, T1 and T2, formed by joining wire types A and B.1

The ice/water bath at 0°C (Figure 2) has become the standard for the reference

temperature, and published thermocouple voltage vs temperature tables are based on

1 Figures are from Analog Devices Application Note AN-369, Thermocouple Signal Conditioning Using the AD594/595, J. Marcin, 1998.

 

 

8/17/16 page 2 of 7

this value. This method of providing the reference junction temperature is impractical in field- and lab-measurement situations, so alternatives have been developed.

 

Figure 2. Thermocouple loop with the reference junction at 0 °C.

Rather than using ice, two methods can be used to do cold-junction compensation

– software and hardware. The temperature of the reference junction can be measured directly using a semiconductor sensor or thermistor. The T2 sensor can be chosen to provide a very accurate measurement in a narrow temperature span centered on the expected junction temperature. The measured T2 and the measured sensing junction voltage can be used in a calculation to remove the effect of the reference junction voltage and extract the temperature of T1.

The alternative to the software approach is to have the T2 sensing junction within

your measurement hardware, and have it used by a circuit that will generate a voltage equal and opposite to that of the reference junction (Figure 3). Once the effect of the T2 junction is removed, the circuit amplifies and scales the output voltage to represent the T1 junction temperature as 1 mV/°C or 10 mV/°C (the 10 mV/°C value is more common).

 

Figure 3. Electronic cold junction compensation

 

 

 

ENGR 202 Evaluation and Presentation of Experimental Data II – Summer 2016 Lab 4: Capturing Temperature Measurements with a

Thermocouple Original: Dr. Scoles, Dr Miller, Dr Chmielewski Rev: Dr. Marino

 

8/17/16 page 3 of 7

Procedure

1. With the Hewlett Packard E3631A DC power supply off connect two alligator leads from the power resistor terminals to the + and COM terminals (under ±25V label) as shown in Figure 5).

 

Figure 4. TC inserted into the core of the bower resistor (not to scale)

 

1. Start your temperature measurement VI. Within the NI software, set the thermocouple type to, ‘K’ and set the units to ‘C’. Enable data logging in the NI software, collecting 1 sample/second.

2. Record the starting temperature as the ambient temperature, T∞, in the analysis discussion that follows.

3. Insert the Omega thermocouple into the center of the power resistor. The thermocouple should not touch the sides of the resistor, it must float at the center of the radius.

4. Set the power supply voltage to 16 V.

a. Turn on power supply by pressing the “Power” button. Press the “Output On/Off” button. Voltages are shown on the left half of the supply display, and currents on the right half. If a digit on the voltage side is not flashing, press the “Voltage/Current” button. Use the “Adjust” knob to set the voltage.

5. Observe the TC temperature as the resistor warms up to its maximum temperature, typically between 50 and 70°C (122 and 158°F). Collect data until dT/dt=0.0167, or one degree/minute, we will treat that as steady-state.

6. Remove the TC from the power resistor, hold it vertically without waiving it around and continue to measure the temperature until it returns to a value close

 

 

8/17/16 page 4 of 7

to the ambient level. This is measuring the free convective cooling response of the thermocouple.

7. Stop your VI and save the data to a file.

8. Repeat this heating and cooling cycle two more times, saving the data into a new file each time.

9. Make sure you have three good cooling curves saved before you leave the lab. These curves should generally look alike.

10. Turn off the power supply.

a. Press “Output On/Off” on the supply, and turn off “Power”.

Data Analysis – this portion can be done outside of lab Part 1.

The shape of the curve you saw for the thermocouple cooling is characteristic of many physical phenomena, including capacitor discharging, radioactive decay, and others. A straight forward energy analysis of the thermocouple system identifies that the rate of change in energy stored in the thermocouple is equal to the energy lost to the room by way of convection.

The energy of the system is calculated with respect to the heat capacity of the thermocouple and is represented by the expression

E = m x cp x T (1)

Where E = energy content of the thermocouple, kJ m = mass of thermocouple system, kg (assumed constant) cp = the specific heat of the material from which it is constructed, kJ/(kg-K)

(assumed constant) T = temperature of the thermocouple, K, which varies.

Therefore the rate of energy change with respect to time is evaluated by taking the time derivative of this equation

dE/dt = m x cp x dT/dt (2)

Where t is time in seconds. The energy leaving the thermocouple is picked up by the air in the room. This

energy flow, driven by the temperature difference between the thermocouple and the air in the room is called heat transfer and in this case is primarily convective heat transfer (we will ignore conduction and radiation). As mentioned in lecture, this mode of heat

 

 

ENGR 202 Evaluation and Presentation of Experimental Data II – Summer 2016 Lab 4: Capturing Temperature Measurements with a

Thermocouple Original: Dr. Scoles, Dr Miller, Dr Chmielewski Rev: Dr. Marino

 

8/17/16 page 5 of 7

transfer is modeled based on the Newton Law of Cooling for a surface and is calculated with the expression

dE/dT = h x As x (T-T∞) (3)

Where h = Newton Coefficient for rate of convective heat transfer, kJ/(m2-K-s)

depends on the conditions As = surface area of the thermocouple, m2 T∞ = temperature of the room, K (this is the ambient temperature of the room) T = temperature of the hot surface, K, in this case the thermocouple

temperature

Equating the two expressions for rate of energy change produces a simple, first order ordinary differential equation between temperature and time

dE / dT = – m x cp x dT/dt = h x As x (T-T∞) (4) Note: the negative sign results from the fact that energy gain by the air is energy lost from the TC Take a look at the simple solution for this equation of temperature as a function of time, T(t). Determine the time constant, τ, for a first-order thermodynamic system:

Where T0 = temperature of thermocouple before cooling starts, K

 = m cp / (h As), s, represents the time constant for the first order system

T∞= ambient temperature

The next formal step is usually to collect terms in T and t, which yields

With the data acquired in this laboratory, T∞, T0, and T(t) the time constant for your TC can be evaluated. There are several ways to find these time constants.

    (5) /t 0

e TTTtT 

 

 



T t T T

0 T



e t /

(6)

 

 

8/17/16 page 6 of 7

The simplest technique is to take the natural logarithm of both sides of equation 1, which yields

 

This equation has the familiar form of y = mx + b, where the slope m equals -1/ and the intercept b is 0. The slope of the straight line you get when you plot the natural log of the fraction in parenthesis vs. time will be the time constant. The Excel LINEST function can extract the slope from straight-line data.

Computer tools such as MATLAB and LabVIEW have built-in capabilities to fit an exponential curve to a set of data. See the Exponential Fit VI in LabVIEW’s Mathematics: Fitting menu on the Functions Palette. Tab-delimited data can be read into LabVIEW with the Read From Measurement File Express VI.

Once you find your time constant, plot an exponential through your measured data. Describe in your report how well an exponential model fits the cooling data.

For each run, calculate:

o Rise time to steady state and dT/dt at mid-rise

o Time at steady state and dT/dt

o Fall time from steady state and dT/dt at mid-fall

Your Report

Prepare a written report following the guidelines in our grading rubric. This report is due one week after your lab. Required Graphical Results

Each of the three cooling curves should be included in the report Some Discussion Points That Must Be Covered

 From your readings and lecture, what are some of the advantages and disadvantages of using the Cold Junction Compensation circuit for temperature measurement?

 Why would you choose a differential input channel thermocouple for this application rather than a single-ended channel?

 Explain your reasoning behind the setup of your voltage measurement task.

 What are some of the sources of measurement error in this experiment? What is the Omega Type K thermocouple temperature accuracy? Can you use propagation of error to estimate the error in the temperature readings?

 How well did the exponential cooling model fit the temperature data?

 



ln T t  T T

0  T





 



  – t/ (7)

 

 

ENGR 202 Evaluation and Presentation of Experimental Data II – Summer 2016 Lab 4: Capturing Temperature Measurements with a

Thermocouple Original: Dr. Scoles, Dr Miller, Dr Chmielewski Rev: Dr. Marino

 

8/17/16 page 7 of 7

Required Printouts

 One page hardcopy of final temperature measurement front panel showing measured data for the three trials

 Excel, Labview, or Matlab analysis of your temperature measurements. Make sure all tables and figures are properly labeled in the body of your lab report or the appendix section.

 If you use Labview: one page hardcopy of final temperature measurement block diagram including the subVI. The diagram should have text documentation explaining the VI’s function and the team member names.

Bibliography

American Society for Testing and Materials (ATSM), Manual on the Use of Thermocouples in Temperature Measurement, ASTM PCN 04-470020-40.

Analog Devices AD594/595 Datasheet, http://www.analog.com/UploadedFiles/Data_Sheets/AD594_595.pdf, Rev. C, 1999. Viewed on November 3, 2007.

Omega TC wire spec sheet page

Analog Devices Application Note AN-369, Thermocouple Signal Conditioning Using the AD594/595, J. Marcin, 1998.

Doering, Ed. Create a SubVI in LabVIEW, http://cnx.org/content/m14767/latest/ Connexions. 17 Mar. 2008. Viewed on May 10, 2010.