This coil, then, with the thermo-couple at its lower extremity, is hung between the ends of a powerful magnet much as the fibre of the Einthoven Galvanometer is situated. The alternating current to be measured comes along through the heater. The heater rises in temperature. That warms the lower end of the thermo-couple. Instantly a steady, continuous current begins to circulate round the silver strip which forms the coil, and that, acting just as the current does in the ordinary galvanometer, causes the coil to swing round more or less, which movement is indicated by the spot of light from the mirror. A current as small as twenty micro-amperes (or twenty millionths of an ampere) can be measured in this way.
Mr Duddell has also perfected a wonderful instrument called an Oscillograph, for the strange purpose of making actual pictures of the rise and fall in volume of current in alternating circuits.
Fig. 3.—The "Duddell" Thermo-galvanometer. In this remarkable instrument alternating current enters at a, passes through the fine wire and leaves at b. In doing this it heats the wire, which in turn heats the lower end of the bismuth and antimony bars. This generates continuous current, which circulates through the loop of silver wire, c, which, since it hangs between the poles, d and e, of a magnet, is thereby turned more or less. The amount of the turning indicates the strength of the alternating current.
To realise the almost miraculous delicacy of these wonderful instruments we need first of all to construct a mental picture of what takes place in a circuit through which alternating current is passing. The current begins to flow: it gradually increases in volume until it reaches its maximum: then it begins to die away until it becomes nil: then it begins to grow in the opposite direction, increases to its maximum and dies away once more. That cycle of events occurs over and over again at the rate it may be of hundreds of times per second. Now for the actual efficient operation of electrical machinery working on alternating current it is very necessary to know exactly how those changes take place—do they occur gradually, the current growing and increasing in volume regularly and steadily, or irregularly in a jumpy manner? Engineers have a great fancy for setting out such changes in the form of diagrams, in which case the alternations are represented by a wavy line, and it is of much importance to obtain an actual diagram showing not what the changes should be according to theory, but what they really are in practice. It is then possible to see whether the "wave-form" of the current is what it ought to be.
Once again we must turn our thoughts back to the string galvanometer. In that case, it will be remembered, there is a conducting fibre passing between the ends or poles of a powerful magnet, the result of which arrangement is that as the current passes through the fibre it is bent by the action of the magnetic forces produced around it. If the current pass one way, downwards let us say, the fibre will be bent one way, while if it pass upwards it will be bent the opposite way. Suppose then that we have two fibres instead of one, and that we send the current up one and down the other. One will be bent inwards and the other outwards. Then suppose that we fix a little mirror to the centre of the fibres, one side of it being attached to one fibre and the other to the other. As one fibre advances and the other recedes the mirror will be turned more or less. Consequently, as the current flowing in the fibres increases or decreases, or changes in direction, the mirror will be slewed round more or less in one direction or the other.
The spot of light thrown by the mirror will then dance from side to side with every variation, and if it be made to fall upon a rapidly moving strip of photograph paper a wavy line will be drawn upon the paper which will faithfully represent the changes in the current.
In its action, of course, it is not unlike an ordinary mirror galvanometer, but its special feature is in the mechanical arrangement of its parts which enable it to move with sufficient rapidity to follow the rapidly succeeding changes which need to be investigated. It is far less sensitive than, say, a Thomson Galvanometer, but the latter could not respond quickly enough for this particular purpose.