It is common after a days work people take a bath and there are countless occasions the water is not hot enough. Many have often wondered why there is not some kind of product on the market to deal with this problem, something to sense the temperature of the water inside a domestic hot water tank. Aim: To design and test a sensor that would be capable of measuring the temperature of water inside a domestic hot water tank (20-80i?? C). Design of the sensor: To fully grasp the needs of this project it is necessary to understand the workings of a hot water tank. A coil heats the water. Inside the water is in a continuous loop.

The gas boiler heats the water in the loop this then flows from the boiler to a coiled loop inside the boiler. The coil becomes hot and heats the water inside the tank due to the second and zeroth law of thermodynamics (i. e. Heat flows spontaneously from a hotter to a colder object but not vice versa. Because of this, it is impossible for a cyclical system to transfer heat from a lower-temperature body to a higher temperature body indefinitely unless external work is done on a system). To heat the water even faster a powerful electrical heater is inserted into the boiler which heats water faster, an immersion heater.

A standard thermostat, which ensures no overheating of the water tank, relies on a bi-metallic strip. This is built on the principle that not all metals expand at the same rate and bends when it is hot. The bi-metallic strip is not suitable however for the purpose we require. We require an accurate reading of the temperature to ensure a hot bath is achieved and also the sensor could act as an energy saving device stopping unnecessary over heating of the water inside. An accurate way to do this would be to utilise a device, which varies electrically with change in temperature such as a thermistor.

A thermistor generally are small beads of metal oxides which are surprisingly good insulators because they lack the sea of free unpaired electrons that the metal has in which the electricity is conducted. Their conductance (Conductance = 1/Resistance) increases rapidly with temperature and resistance decreases. Circuit Design These very small thermistors require a very small current running through them as if there is too much this will cause self-heating of the sensor which will render the results and ultimately the sensor useless. Typically under 100i?? A avoids self-heating and does not drain the battery.

After researching the best circuit set up for this sensor I was left with two options. The first, a potential divider, (Fig2) would give voltages readings with a voltmeter set up in parallel. The other type of suitable circuit architecture I could use is a “Wheatstone Bridge”. It is mainly used to sense very small changes. When popularised by Sir Charles Wheatstone he used a very sensitive device to measure current called a Galvanometer. We however have access to cheap but accurate digital multimeters, which can measure to this degree accurately anyway and it is then possible to measure either potential difference or current.