2 Getting into hot water
2.1 Boiling water
Whether it's to wash clothes, make a cup of tea, or just make it safe to drink, water often has to be heated – sometimes to boiling point. There are many ways to do this, but a very common means is some form of electric water-boiler, such as a kettle or an urn. In all but the crudest ones, a device is fitted to ensure that heating does not continue once the boiling point of water is reached.
In deciding on the type and design of such a device, we can suppose that a company manufacturing electric water-boilers would have gone through a process similar to the solution-finding methods described in OpenLearn course T207_1 The engineer as problem-solver: the nature of problems. Recalling from there the diagram of the decision-making route (which is attached below for ease of reference), the need is for electric water-boilers that have a device fitted to ensure that the water does not continue to be heated once it has reached boiling point.
Click on the 'view document' link below to view the diagram of a decision-making route
View document [Tip: hold Ctrl and click a link to open it in a new tab. (Hide tip)]
The next step is to think more deeply about the need and decide whether there is more to it than this. For example, it has been taken for granted that the need is for this device to work by itself. In other words, we would not be satisfied with a whistle that merely reports that the water is now boiling. So to make this part of the need explicit, the statement can be modified to say that the device must cut off the power automatically once the water has boiled (Figure 2).
Example 1
List five other aspects of the device that could be included to further define the need.
Answer
My list looks like this. It's already more than five items, and with a bit more thought it could be lengthened. You may have thought of some of these additional items.
The device should work by cutting off the current.
It should be triggered accurately near the boiling temperature.
It should be re-settable once triggered (i.e. no irreversible changes).
It should be reliable.
It should be robust enough to work repeatedly over a long time without requiring maintenance, adjustment, or replacement of any parts.
It should be cheap.
It should be small enough not to add significantly to the size of the boiler.
It should be easy to use.
Our responses to the preceding example begin to give a fairly detailed picture of the essential features of the solution being sought. Parts of the list can already be grouped together and called a functional specification. This includes such items as whether or not the device should be re-settable. If it is to be, this can be further defined by stating whether the re-setting is to be automatic or manual. If automatic, once the water has boiled, it would be allowed to cool by only a certain amount before the power is re-applied (i.e. a thermostat). If manual, it would let the water cool right down again (i.e. a cut-out). Such questions can only be answered by referring back to the final user, and are one reason for the existence of marketing departments.
The functional specification so far just itemises what the solution should do, but it says little about how well it should do it. To make the specification a true yardstick against which the solution can be measured, we need to quantify things.
With numbers against each quantifiable item on the functional list, it becomes a performance specification. It may state, for instance, that the device must operate when the water reaches a temperature within the range 97 °C to 100 °C. Much of this information again comes from the user, and it is surprising how flexible and arbitrary many of these numbers can be. For instance, thermometers for the control of domestic ovens could be specified with an accuracy of ±0.1 °C of reading, and I daresay you'll find someone who'll install one and charge accordingly, but ±5 °C is good enough. Specifications should be prepared and read very carefully, because it is not always easy to distinguish between the 'wish list' or 'guesstimate' numbers and those which really must not be exceeded. The example of the operating temperature of a water-boiler serves to illustrate the point: the 97 °C figure is more flexible than the 100 °C. If the lower limit were reduced to, say, 92 °C, the result would merely be water a little further below boiling point than otherwise. If, however, the upper limit were relaxed to 105 °C, the device may never be triggered as the water will boil away before it reaches that temperature, rendering it (and eventually the boiler itself) useless.
Armed with this quantitative data, the time has come to generate some ideas. But wait! The engineers can't do this without having in their heads a store of knowledge and understanding of how things work. A mark of a really good engineer is an interest in a wide range of technologies and sciences. This enables them to draw analogies, to transfer ideas from one context to another, and to make rapid but accurate predictions of what will and won't work. It is this kind of holistic approach that enables the innovations by context covered in OpenLearn course T207_1 The engineer as problem-solver: the nature of problems to come about. For the water-boiler device, what is essential is knowledge of how temperature affects different materials and material properties. From this we can begin to propose effects to be exploited in a thermal cut-out.
I want to rule out the use of the approach of using a routine solution at this stage. Of course one could just find a catalogue or website that lists thermal switches that are already on the market. Instead, I am encouraging an innovative approach – and that calls for a deeper knowledge of the effects of temperature.
In Sections 3, 4 and 5, we'll look at temperature effects in detail. Afterwards you will find yourself equipped not only to produce more ideas when you return to the water-boiler, but also to take the next step of evaluating these ideas in terms of their appropriateness for this application.