Experimenting with Thermocouples.
Extracts from this essay...
Physics AS Level Coursework Experimenting with Thermocouples Introduction For my sensor coursework, I have chosen to investigate the properties of thermocouples. A thermocouple is a sensor which detects a temperature difference, and produces a very small electrical output. "In 1822, an Estonian physician named Thomas Seebeck discovered (accidentally) that the junction between two metals generates a voltage which is a function of temperature. Thermocouples rely on this Seebeck effect. Although almost any two types of metal can be used to make a thermocouple, a number of standard types are used because they possess predictable output voltages and large temperature gradients." Source: http://www.picotech.com/applications/thermocouple.html Welding, or otherwise combining, two dissimilar metals can make a thermocouple. Varying the temperature of the junction where the two metals combine will produce a very small voltage and a very small current. However, if one attempts to connect the thermocouple to a Voltmeter, another thermocouple junction is made. This is at the point of contact, where the ends of the thermocouples meet the contacts of the Voltmeter, and causes problems, as it can lead to errors in the result. To compensate for this, a technique known as cold junction compensation (CJC) is used. This entails adding an extra wire of the first material at the end of the thermocouple, so that the metal in contact with the voltmeter are both made from the same material, which cancels out the potential difference. A result of the CJC is that there are two free junctions. One junction is kept at a constant temperature for the other junction to refer to. That is why this junction is commonly known as the "reference" junction. As most reference junctions are kept in an ice bath at a stable temperature of 0 oC, it is also known as the "cold junction", hence the term "cold junction compensation". Accordingly, the other junction that actually measures the temperature is commonly referred to as the "hot junction".
As previously, as the temperature decreased, the reading decreased at the same rate. I then put some crushed ice in the boiling water. This caused the water to cool far more rapidly. I noticed that this time, the Galvanometer was fluctuating considerably, and only stabilised when the ice had completely melted, thus restoring the cooling rate of the water to normal. Determining the Sensitivity a Thermocouple To gather suitable results, I varied the temperature difference between the junctions (independent variable), whilst measuring the current flowing through the wire (dependent variable). To vary the temperature difference, I placed the "cold" junction in an ice bath to keep it at a constant temperature of 0 oC. N.B The ice bath can be re-used from the calibration of the Galvanometer. I then immersed the "hot" junction in boiling water. As the water temperature decreased, the temperature difference decreased from 100 oC down to room temperature. As cooling began immediately, I began observations at 80 oC; due the to the fact that the cooling rate decreases as the temperature difference between the hot water and the air decreases, I decided to end observations at 40 oC. It was difficult to gain results from temperatures lower than 40 oC as ice was needed to cool the liquid further. I recorded the reading from the Galvanometer every 5 oC to give me a wide range of results. Method * Clear work area, set out equipment, ensure Galvanometer is set to "x 1" and plugged into the mains. * Attach the Copper-Constantan thermocouple to the Galvanometer, calibrate using method described above * Boil water using kettle, once boiling, pour into insulated beaker and immerse the hot junction, taking care not to insert more than just the junction. * Fill next insulated beaker with crushed ice (or use beaker filled with ice from calibration), then insert cold junction, again taking care not to insert more than just the junction.
When I tested my Iron-Copper Thermocouple, the Galvanometer did fluctuate slightly at first, at which point I sought advice from my Physics Tutor. I was informed that the fluctuations were caused by bad connections between the thermocouple and the Galvanometer. As I was using enamelled copper, I thought it would be wise to scrape of some more of the enamelling around the parts of the wire in contact with the Galvanometer. This fixed the problem, and there were no more fluctuations. There were some elements of the experiment that were beyond my control. Firstly, to ensure a fair test, the lengths of the wires used to construct the thermocouple should be the same for all the thermocouples tested, as different wire lengths would have different resistances, which would affect the current measured. Secondly, the thermocouple junction sizes should also be the same, as a larger junction size would produce a larger potential difference, as there are more electrons available to move in between the metals. A larger potential difference would result in a larger current. Applications Within Industry Thermocouples are already used widely in industry. This is due to a variety of factors, including the cheap cost (excluding Platinum-Rhodium thermocouples), the robustness of the sensor (made from metal wires) and the simplicity (temperature difference will produce a voltage and current through the wires. A practical example for use of a thermocouple is in an old style boiler. When the pilot light (essentially a flame) is turned on, it heats up one junction of the thermocouple. This produces a small current, which is used as feedback to keep the pilot light alight. From my experiments, I can see that there are some drawbacks, however. In Copper-Constantan thermocouples, the smallest temperature change that can be sensed is 1 oC. In some precision measuring applications in industry, this may simply be not accurate enough for the task at hand, such as monitoring a sick person's body temperature in a hospital, for example, as this would deal in looking at changes in temperature in 0.1 oC. Also, it appears that a rapidly changing temperature would cause problems.
Found what you're looking for?
- Start learning 29% faster today
- Over 150,000 essays available
- Just £6.99 a month
- Over 180,000 student essays
- Every subject and level covered
- Thousands of essays marked by teachers