A problem with working with thermocouples is that it needs a sensitive voltmeter, able to detect a signal of the order of a thousandth of a volt. Too sensitive a detector, and there appears to be no temperature difference. Another problem is the problem of internal resistance of a source. If the thermocouple is connected to a low resistance detector, the potential difference across it may be tiny, because most of the potential difference available is used up in driving a small current through the large internal resistance. Only if the detector has itself a very high resistance is there an appreciable potential difference seen as the output of the sensor.
A thermocouple thermometer can be made using two similar wires and a different wire connected together to form two junctions acting in opposition to each other. The free end of each similar wire is connected to a microvoltmeter or to a d.c. Amplifier with a voltmeter connected at its output. One of the junctions is maintained at a constant temperature and the other junction is used as a temperature prob. The meter reading changes according to the difference between the temperature of the probe and the reference junction. The probe is simply a junction between two thin wires and this makes it respond rapidly to change of temperature and has therefore a faster response time in liquid – glass thermometer. In addition, because a thermocouple generates an emf directly, it is used widely in control systems.
The diagram below shows an idealised thermocouple. As you can see it shows a series circuit set up, consisting of a digital voltmeter, two wires of metal x and a single wire of metal y. The junctions at either end of metal Y labelled A and B are set in two objects of temperatures. Oa and Ob respectively. The diagrams show that the reading of the digital voltmeter depends on Oa – Ob. If Ob is fixed at 0’C by immersing the end at b in iced water, the digital voltmeter can be calibrated to give a direct reading of temperature at Oa.
Thermocouples are widely used because of their many valuable qualities:
- They are accurate, rugged and very sensitive
- They are usually small enough not to interfere with the object whose temperatures are being measured
- The output can be fed directly to a data-recording device.
- The instrumentation can be remote from any hot object under investigation
Thermocouples are made by joining two ends of one lead of iron (Fe) wire to two long leads of copper wire. This is done by first of all cleaning the ends thoroughly and then either twisted or soldered together.
Thermocouples are not easy to build because their calibration depends on too many factors such as the purity of the metals used for the wires and the quality of the junctions. The metals used often need special solider to hold them tightly together
Plan
First of all I am going to collect all my equipment that I need in order to carry out my investigation. Below is a list of all the requirements needed.
Requirements
Copper lead*2
Iron lead*1
Sandpaper
Insulating tape – to hold wires together
Wire cutters _to cut wires to desired length
Two 250ml beakers
Ice
Kettle – for hot water
Digital thermometer – to measure temperature
Micro voltmeter
12v power supply – for power
First of all I will need to make my thermocouple. I will simply do this by cutting two equal lengths of some wire, the copper wire and one length of the other iron wire. I will then assemble them together by first of all cleaning the ends using sandpaper and then twisting the ends together of the different metals to make the two junctions. I will then secure the ends together using insulating tape around the wires to prevents them disconnecting when in use. I could alternatively solder the ends together, however this can be difficult to do to get a secure joint, so ill do it this way which is just as effective.
I will then need to connect the micro-voltmeter to an ac power pack supply and then connect the micro-voltmeter to both ends of the thermocouple. I must make sure that my results are in milli-volts so I must change my range on the micro-voltmeter to milli-volts.
Next I will cool some cold water with ice in a beaker, getting it near enough 0’C as I can, and stick both my junctions into it. I will then zero my micro-voltmeter, because both my junctions are at the same temperature, therefore it shouldn’t give me an emf.
Now I will be ready to begin my experiment by keeping one junction at 0’C that is my reference temperature, and warming the water at the other junction to vary the temperatures. I will then record my result. First I will do a trial to see if the thermocouple works, then I will repeat my results three times in order to get an average of my results.
DIAGRAM
RESULTS
Below is a table of my first results that I obtained for my trial investigation
These that I obtained from my first experiment are very pleasing results, which shows me that my thermocouple does work. However I did find a few problems whilst I was carrying out my experiment.
First of all while I was taking my initial results I discovered that there was quite a lot of noise. I found that if I moved around or any slight movements around where my experiment was set up, then this created more noise. I resolved this problem by moving to another location of the room and kept clear of any distractions, which helped a lot.
I also had to take my results from 3’C, as I couldn’t get the water to cool below this point. However this is significant as I will be taking the results from this point throughout my investigation and taking an average from this as well.
I could also only get up to 83’C maximum as I only had access to a kettle and not a Bunsen burner. But this is also insignificant and wont be much of a problem.
I can now get on with my experiment knowing the results I should be getting.
Here are the results that I obtained after repeating my results with my improved thermocouple.
Results table 1:
I have repeated my experiment to confirm accuracy and reliability of my evidence. This is shown in results table 2 and once again in results table 3.
Results table 2:
Results table 3:
From these results you can see that the results that I have obtained are quite similar but are not exact each time. To over come this I have worked out the average of my results by adding up all the values for each temperature range from tables 1, 2 and 3, and dividing the value by three. The average results are shown in table 4.
Results table 4:
See graph 1 for graphical representation of my results.
ANALYSIS
From the graph you can see that the relationship is linear. The straight line represents this relationship, which means that E, e.m.f., is proportional to T, temperature. This can be written in the form y= mx + c, where y is the emf, x is the temperature, m is the gradient of the line and c, is the intercept on the y-axis. This gives us a general formula to approximate the output for a given temperature
