Thermistors are thermally sensitive resistors and have, according to type, a negative (NTC), or positive (PTC) temperature coefficient. They work because at low temperatures, electrons are fixed onto atoms and so cannot move. As the electrons get hotter they receive enough energy to break away from their atoms, so the thermistor becomes a better conductor.
In this experiment I do not know if I will have a NTC or a PTC thermistor but if the resistance decreases as the temperature increases then I will know I am using a NTC thermistor, if the temperature and resistance are opposite to that of a NTC then I will be using a PTC.
NTC thermistors offer many features for temperature measurement and control within their operating temperature range. NTC are also known as a ceramic semiconductor.
NTC thermistors show a decrease in electrical resistance with increasing temperature. Depending on the materials and methods of manufacture, they are generally used in the temperature range of –50 degrees to 150 degrees, and up to 300 degrees for some glass units. The resistance value of a thermistor is typically referenced at 25 degrees.
Thermistors are semi conductors so the resistance of a thermistor varies with temperature. A thermistor can be considered to be a semi conductor therefore the resistance across it will decrease as the temperature increases the resistance across it decreases.
Semiconductors are solid materials with conductivities in between the very high conductivity of metals and the very low conductivities of insulators. There are a variety of types of semiconductor, including metal oxides as well as elements like silicon. In insulators, all the electrons are tightly bound to atoms or ions, and none are free to move under an external electric field. Therefore these materials do not conduct electricity at all. In metallic conductors, essentially all the atoms are ionised, providing free electrons, which move freely through the ions and can move under an external electric field. These electrons 'glue' the ions together, and provide non-directional bonding which holds the material together. They become shared amongst all the atoms in the material instead of remaining attached to one atom.
Semiconductors differ from both insulators and metallic conductors. Only a small proportion of atoms are ionised, so that although there are conduction electrons they are relatively small in number and the material conducts, but not well. At higher temperatures, more atoms are ionised, and the conductivity rises.
Apparatus:
Diagram:
Method:
- Collect and set up all apparatus as shown
- Record the resistance of the thermistor at room temperature.
- Place ice in beaker to get to lower temperatures.
- Wait for temperature to get as low as 10 degrees Celsius
- Place thermistor in the beaker of water
- Record resistance from voltmeter.
- Remove ice from the beaker and switch on Bunsen burner.
- Allow the temperature to increase
- Record results in 10 degree intervals up until 100 degrees Celsius
- Repeat steps 3 – 9 twice for repeats.
- Work out average and plot calibration graph whit resistance against temperature.
Risk Assessment/safety:
There is only element that provides a major risk in this experiment this being the extremely hot water. There is no real protection against this but if precautions and care is taken with regards to the surroundings and others working in the vicinity. If scalding occurs then the skin will have to be submersed in ice water to minimise the pain. Although water and electricity are not usually a good combination the fact that only 5V are passing round the circuit there is no real danger from the charge.
Fair test:
In order to make this experiment a fair test I will have to perform the experiment under the same conditions as each other and be as precise with my results as possible. To do so I will write my results correct to the 3rd decimal place, which is rather precise. Also I will try and conduct them under the same room temperature as the slightest of alteration could affect my results considerably. I will conduct repeats in order to make my results more reliable. I will leave the thermistor in the certain degrees of water for the same amount of time each time when recording my results as sometimes it may take longer for an accurate reading to be taken down and it takes time for the thermistor to get used to the new temperature.
A couple of other important things were ensured in order to keep the results fair and accurate. One was to keep the thermistor as close to the thermometer's bulb as possible. This was so that their readings were being affected by the same part of the water because (due to convection currents and uneven heating) the temperature of the water was not equal throughout the beaker.
Results:
Conclusion:
Firstly I found out that I was using a NTC thermistor as I found out that as the temperature of the water increased the p.d across the thermistor decreased. This is expected with a NTC thermistor. This is due to the semiconductor material of the thermistor. The material becomes more conductive as the temperature increases. The reason for this is because some of the electrons of the semiconductor are able to cross from the valance band to the conductance band. The sensor performed relatively well as only one anomalous result was found, this was at 10°C, the reason for this was maybe that the thermistor could not perform as accurately under such cold temperatures as for all the other readings the thermistor seemed to be working in perfect order. The resistance at different temperatures were very similar all three times I took a set of readings, when the temperature was increasing and when the temperature was decreasing. The sensor performed as expected and produced very accurate and precise results correct to the third decimal place.
I plotted my results on a graph where the shape of a straight line appeared which indicated that there was no real fault or indication that my thermistor was not working properly. All results came up as they should have been coming up. I calculated the gradient of the straight line at 0.03, which indicates that it was quite sensitive and could detect small changes in temperature.
Therefore I can conclude that my thermistor performed fairly well, this is because the results followed a pattern, my thermistor performed successfully because the response time was good, there was an unsystematic error and therefore I can conclude that my readings were fairly accurate
Evaluation:
There was an anomalous result at 10°C. This is due to a number of reasons described below. For these reasons, this system will not be useful where extreme accuracy is required, but it was adequate for the proposed application.
Below I have described the factors, which will have affected the accuracy of the thermistor's performance.
I chose water as convection occurs quickly and the heat is fairly evenly distributed. I also made sure that the thermometer and thermistor were very close to each other. However there still may have been some slight differences in temperature between the point, which was being measured by the thermometer, and the thermistor. Therefore, next time it would be a good idea to stir the water constantly.
Further more the voltmeter may have needed to settle, but because the water was always being heated or cooled, it would have not had time to do so and so was not reading the resistance/voltage at the actual temperature, but rather it was behind.
In addition to this the thermistor itself has a relatively slow reaction time to the change in temperature. By using a resistance thermometer, this response time can be decreased dramatically. The thermistor may have been releasing some energy also which it will have then sensed, and will have affected the thermistor's resistance.
The thermometer has a 0.03% error margin and the thermistor has a 10% error margin. This gives a total maximum percentage error of 10.03%, which is very high. The accuracy and stability of the voltmeter is an additional consideration to add to this figure. Such a high percentage error margin is likely to have been the most influential source of inaccuracies in the results.
Further more I could do other investigations having done this in the future I could use a PTC thermistor instead of the NTC thermistor used. Also I could use other semi conductors or even I could change the resistance to see its effects on temperature, I could also go back to my original idea, which was to see what effect the temperature of water has on a coiled metal wire.
