ENGG252 Lab4 Report PDF

Preview

Summary

Full marks for report....

Description

LAB 4 Report: Temperature Measurement 1.

Aim The aim of this experiment is to provide students with an introduction to temperature measuring equipment and techniques. This includes construction, calibration and transient response of sensors.

2.

Introduction The measurement of temperature using thermocouples relies on the Seebeck Effect, discovered by Thomas Seebeck in 1822. This consists of two thermo-junctions, which are made by joining different conductive materials at one end. At a given temperature electrical potential is generated at the junction of leads. This voltage is shown to be directly proportional to the temperature difference between the junctions and the Seebeck (thermocouple) coefficients (units of mV/℃), for that particular pair of metals. The purpose of this experiment is to observe the Seebeck effect to understand the behaviour of thermocouples. This is achieved through the study of thermocouple construction and experimenting with these thermocouples in two parts. The first involves calibration of several temperature sensors against a mercury-in-glass reference thermometer. The second involves assessment of the transient response of the sensors to the step change in temperature (from room temperature to 50°C) as functions of time. Students are able to observe results from LabView software to calculate the Seebeck and platinum resistance coefficients, and then determine the temperature outputs of voltage and resistance sensor outputs, using the following equations:

ΔV = αΔT Equation 1: Seebeck Effect Relationship where V = Voltage, T = Temperature and α = Seebeck Coefficient

R = Ro (1 + β ΔT ) Equation 2: Resistance Coefficient where R = Resistance, Ro = Initial Resistance, T = Temperature and β = Resistance Coefficient

In analysing the transient response of sensors, students assessed the approach of the sensor temperature to the final water bath temperature as being approximately exponential according to the relationship provided in the Laboratory Manual:

T predicted = (T start − T f inal)e−t/Ω + T f inal Equation 3: Transient Sensor Response Relationship where T = Temperature

Plotting the above results on graphs allowed students to visualise the relationships established in Equations (1) → (3). The analysis of thermocouple behaviour has its uses in approximating temperature differences or actuating electronic switches of large systems. The industrial uses for this capability exist in thermoelectric cooling technology.

1

3.

Procedure 3.1. Calibration of Sensors Table 1: Sensor types to be tested (from Lab Manual)

Figure 1: Schematic of Experimental Apparatus (from Lab Manual)

a) The thermocouples and platinum resistance thermometer were purchased or constructed prior to the experiment. The lab supervisor explained the construction of each thermocouple (Type T copper/constantan, Type K chromel/alumel) as well of the experimental apparatus and data acquisition system, LabView. Each water bath is heated to a different temperature reading according to a mercury-in-glass thermometer. These temperatures are 25 °C in the first bath and 50 °C in the second bath. b) Recording with the LabView program, the sampling rate is set as 10Hz and all the sensors are placed in the ice-water reference bath. The respective readings from LabView are recorded in the same table. c) The sensors (excluding the cold junction of our thermocouple pair) are then placed in the first calibration water bath (25 °C). After voltage and resistance readings have stabilized, the respective readings from LabView are recorded in our laboratory notebooks. d) The sensors (excluding the cold junction of our thermocouple pair) are then placed in the second calibration water bath (50 °C). After voltage and resistance readings have stabilized, the respective readings from LabView are recorded in our laboratory notebooks. 3.2. Assessing Transient Response a) The temperature sensors are removed from the water bath after completing Part 3.1, dried and then allowed to stabilize at the ambient room air temperature. b) After setting the sample rate to 1000Hz, recording is started and the sensors (excluding the cold junction of our thermocouple pair) are plunged simultaneously into the second water bath at 50 °C. When the temperatures have stabilised (after a few seconds) the data logging is stopped. The electronic data from LabView is then imported into Excel/Google Sheets and the data is saved to a personal storage device. 4.

Results 4.1. Calibration of Sensors The results from completing the procedure in 3.1 were recorded in the following table: Table 2: Results from Calibration of Sensors Mercury-in-glass thermometer

Sensor 1: Type K

Sensor 2: Platinum Resistance

Sensor 3: Type T thermocouple

Sensor 4: Type T thermocouple with

2

thermocouple ( )

Thermometer (Ω)

0

1

0.6

0.00002

100.3

25

26

26

0.001

110.3

50

50

50

0.002

119.3

(

)

(

Ice-Water cold junction (mV)

)

Using Google sheets, the raw data for all four temperature sensors was plotted with respect to the temperature of the reference mercury-in-glass thermometer on the horizontal axis. The individual data points as well as the linear trend lines were illustrated in the following graphs.

Figure 2: Calibration of Sensors 1 and 3 Graph

Figure 3: Calibration of Sensor 2 (PRT) Graph

From these results, the Seebeck coefficient and resistance coefficient (dR/dT) were determined from our Sensor 4 and Sensor 2 results respectively using the equations (1) and (2). ΔV = αΔT Equation 1: Seebeck Effect Relationship Applying recorded results from 0°C and 50°C, (0.002−0.00002) α = ΔV ΔT = (50−0) α = 4e − 8

Figure 4: Calibration of Sensor 4 Graph

R = Ro (1 + β ΔT ) Equation 2: Resistance Coefficient Applying recorded results from 0°C and 50°C, and rearranging for β as the subject: (119.3−100.3) β = ΔTΔRRo = (50−0)100.3 β = 0.00379 K

−1

4.2. Assessing Transient Response The above calculations were then used to determine the temperature readings from resistance and voltage readings from sensors 2 and 4 respectively, by rearrangement so that:

T =

V α

Equation 1: Seebeck Effect Relationship

T =

ΔR βRo

Equation 2: Resistance Coefficient

The transient response of all the sensors to the step change in temperature imposed in section 3.2 was recorded on LabView and plotted below as a function of time, as shown below: 3

Figure 5: Schematic of raw data from a transient temperature test

These results were adjusted to demonstrate the approach of the sensor to the final water bath temperature by adjusting the time of the start of the experiment (for a closer analysis of the exponential curve.

Figure 6: Graph showing least squares best fit to raw data after adjusting time for start of experiment

We had too much difficulty adding series for the second graph, so an average of all four sensors was taken as a function of time and graphed above in Figure 6. The trendline (using the least squares method) demonstrates that the approach of the sensor temperature to the final water bath temperature is approximately exponential. 5.

Discussion 5.1. Identification of any relationships observed in the experimental data. Calculation of Statistical significance of the relationship. In the experiment the relationship that was observed is between the temperature of the water, the voltage, and resistance. The relationships are defined by two equations, the Seebeck Effect Relationship and the Resistance Coefficient. The significance of the relationship is being to use the coefficients to find out the temperature with a known voltage or resistance. 5.2. Comparison of results with other known work. The Experiment conducted can be compared to a similar one conducted by Arman Molki in 2010, which aimed to give a simple demonstration of the Seebeck Effect. The Experiment was similar as the T-type thermocouple is submerged in an Ice bath and boiling water (rather than our 50 degrees) also a control is done with the ice bath and room temperature below is a table comparing the results from their experiment to the results we obtained: Table 3: Calibration results for Seebeck Effect by Arman Molki

4

Respective Temperature (o C)

Type-T Thermocouple (Ours) (mV)

Type-T Thermocouple (Theirs) (mV)

0

0.00002

0

25

0.001

0.00114

50/100 

0.002

0.0023

Although the Experiment offers very little to compare the results that are available are very similar to what was found during this experiment, the major differences is the increased precision in their results compared to what was found except for in the first point of 0oC where their result yielded 0 and the one conducted here found 0.0002mV overall this shows that if these results are correct in comparison than the rest of the experiment would most likely yield similar results. 5.3. Possible sources of error in the experiment include: a) The measuring instrument itself will cause errors due to limitations in digital processing and physical issues. b) A thermocouple junction does not generate any thermovoltage when it is in 0o C, making an ice bath a good reference point. However the ice bath may not accurately provide this temperature exactly as it melts. The use of “cold junction compensation” for thermocouples (as used in sensor 1 and 4) compensates for the missing thermoelectric voltage due to the fact that the thermocouple cold end at the instrument is not a 0°C exactly. This allows our device to use the established thermoelectric voltage tables to determine the temperature at the hot end with high accuracy. c) According to OnSolution, the sensor is one of the greatest sources of error and the most difficult to compensate for. Thermistors and PT100s have an error of approx +-0.4°, Type K thermocouples can have errors of +- 2.2°, Type T thermocouples have errors of +-1°. Calibration is therefore of significant importance to account for the possible individual error of each sensor as well as ensuring that there are no fans or other sources of cooling or heating located near the reference junction that might further impact accuracy. Insulation can also aid in protecting junctions from this. d) There are inherent variations in alloys. Each batch of wires are unique, and as the alloy percentages vary slightly during each manufacturing process, some error in thermocouple accuracy is prevalent. Within standard thermocouples this accounts for 1% error of the actual temperature measurement at the measuring junction. The solution for this is to order thermocouples with special-limit wires: these are manufactured at the highest tolerances to ensure the least impurities and greatest consistency in alloy ratio. 5.4.

5.5.

Table 4: Calculate the magnitude of the random error (including 95% confidence interval) K thermocouple (℃)

T thermocouple (℃)

Sensor 4:Type T (℃)

Sensor 2: PRT (℃)

SD

0.3074

0.5519

0.7882

2.7691

Confidence 95%

0.6149

1.1037

1.5763

5.5383

Table 5: Calculation of systematic error and its statistical significance.

Mean

K thermocouple (℃)

T thermocouple (℃)

Sensor 4:Type T (℃)

Sensor 2: PRT (℃)

47.70

47.50

46.56

44.08

5

Systematic Error

2.3

2.5

3.44

5.92

The significance of calibrate is to adjust the sensor to being able to measure the result closer to the real temperature. 5.6.

The suitability of the equipment and instruments and suggestions for its improvement. The equipment used was suitable for the experiment. LabView unfortunately demonstrated some significant errors in initial readings during record which were omitted from the experiment (results in excess of -(1e6)), however it is also an industry standard program and did provide suitable data for a relatively accurate demonstration of the experiment. Allowing the program to “stabilize” before taking the readings helps in determining a more accurate result. The thermocouples were slightly inaccurate, this could be due to the inherent variation in alloys which can be accounted for by ordering thermocouples with special-limit wires: these are manufactured at the highest tolerances to ensure the least impurities and greatest consistency in alloy ratio. Errors in the sensor readings could be due to inherent inaccuracies in commercial and home-made thermocouples, due to human error or manufacturing error. To compensate students need to ensure that they reduce other possible sources of error and correctly calibrate the sensors before recording. 5.7.

Comments on any unexpected behaviour that may have affected the accuracy of the results. Some Unexpected behaviour could have included the Thermocouples or Thermometers touching one another, another surface such as the casing or sitting out of the water, any of these can easily occur as it can be the slightest touch and quite unnoticeable and would cause error precautions were taken and this it is believed this did not occur. There was an effect on the accuracy of the “Platinum Resistance Thermometer” this may have been due to the length of the sensor and it is quite possible that skin may have come into contact with it at some point throwing the result off by upto 12 degrees. 5.8. Conclusion: A statement summarising the interpretation of the results. This lab helped us improve our temperature measuring technique and calibrating devices. Using four different types of measuring devices we could see how voltage and resistance relates to the temperature using equations. The results showed that the temperature could be reasonably calculated with the equations. 5.9.

References: Acknowledge all external work included in the report, ordered according to sequence of use. 1. ENGG252 Laboratory Manual and Handbook 2. https://sciencestruck.com/seebeck-effect-with-applications 3. http://vlab.amrita.edu/?sub=3&brch=194&sim=351&cnt=1 4. https://searchnetworking.techtarget.com/definition/Seebeck-effect 5. https://www.researchgate.net/publication/228844038_Simple_Demonstration_of_the _Seebeck_Effect 6. https://www.sciencedirect.com/topics/engineering/seebeck-effect 7. https://onsolution.com.au/blog/temperature-measurement-sources-of-errors/ 8. https://blog.wika.us/products/temperature-products/six-common-causes-thermocoupl e-temperature-measurement-errors/

6...