Now in the case of the "shunt" dynamo, unlike the "series", the less the external resistance the weaker will be the machine, since the amount of current flowing in the magnet-winding and the external circuit depends upon their relative resistant-- To see this plainly, if, as in the case of the diagram of the "serics" machine, we cover up the external circuit of the shunt dynamo, and imagine it without any, we shall see that the magnet-circuit alone offers a path back to the armature, and will consequently take all the current it requires; but, on our providing another path, the current in the magnet-circuit will be decreased; hence, the machine is weakened as the external circuit decreases in resistance.

Turning again to the diagram of the series dynamo, if we were to join the brushes across with a piece of copper wire, we should be said to have "short-circuited" the armature, since the current would take a shorter path than that originally provided for it. Now let us see what happens if we short-circuit the terminals, i.e. the two points at which the external circuit commences. We have just seen that the lower the resistance of the external circuit, the stronger the machine. Here is a case where the external resistance is practically nothing, since we have short -circuited it; the machine, therefore, will generate current to excite the- magnets, which will in turn produce more current, and the amount of current produced will, under such conditions, soon burn up the insulation of the armature-winding.

If we short-circuit the terminals of a shunt machine, - that is to say, if we join the two wires across as they leave the dynamo on the external circuit, - the result is very different, since the machine will no longer give any current at all. The reason being that we have also short-circuited the magnet-winding, since both magnet-winding and external circuit emanate from the same points, the brushes: whereas, in the series machine, the current has to pass through the magnet-winding before it meets the external-circuit wires.

3 The Compound Dynamo. In electric lighting, it would be very inconvenient to have a dynamo which, when driven by an engine running at a constant speed, varied its power and pressure according to the quantity of light burning, and yet this is what happens with "series" and "shunt" machines, since the former becomes stronger, and the latter weaker, on an increase of current being required in the external circuit. In order to avoid this inconvenience the compound-wound machine has been designed, which, as its name implies, is a combination of the "series" and "shunt " dynamos, the magnets being wound in both ways, and in such a proportion, that, although the resistance of the external circuit varies, the pressure of the machine remains constant It will be understood from the diagram of this compound machine, Fig. 617, that, on the external circuit decreasing in resistance through more lamps, etc, being turned on, the shunt-winding weakens the magnets, while that of the series strengthens them, and so a balance is kept.

If, with a shunt machine, constant pressure should be necessary, it could be obtained by regulating the resistance of the magnet-winding to suit the resistance of the external circuit. For instance, supposing the external circuit were to decrease in resistance as we have seen, less current would traverse the magnets, and so the machine would become weaker. But if the resistance of the magnet-winding could l»e decreased also, the amount of current round the-magnets would be maintained. To effecl this adjustment of the magnetic winding on a machine giving 100 volts at (say) 1200 revolutions per minute, the magnet-winding should be cut and an adjustable resistance, such as is shown in Fig. 618, inserted; that is to say, the current traversing the magnet-winding would have to traverse the resistance-frame before proceeding through the remainder of the winding.

The resistance in circuit with this winding would depend upon the position of the movable switch-lever, which is so arranged that, by moving it over the different studs on the frame, any number of resistance-springs can be thrown in or taken out at will. If the switch-lever is drawn over to the left as far as the first stud, the whole resistance is short-circuited, since the current enters at the left-hand terminal, and passes, by way of the first left-hand stud, through the switch-lrver and the right-hand terminal, without having been through any of the resistance-springs. If, however, the lever is drawn on to the right-hand stud, the current will have to traverse all the resistance-springs before arriving at the left-hand terminal.

Fig 617. Diagram showing Compound Dynamo and circuit.

Fig. 618- Shunt Resistance-frame.

With such a dynamo and resistance-frame, if the machine is driven at (say) an extra 50 revolutions per minute, making 1250, at which perhaps it would give 105 volts, sufficient resistance should be inserted to reduce this voltage to 100 as originally required. Again, on the voltage dropping through the external resistance decreasing, a little resistance could be cut out to bring it up to 100 again: and similarly, on the voltage increasing through lamps being turned off, resistance could be inserted.

These resistances are frequently used in the shunt-winding of compound dynamos, as well as in plain shunt machines, because, although a compound machine should regulate itself to within about two volts of full load on "open circuit" at a constant speed, yet, in many instances, the speed of the engine cannot be "governed" thus constantly.

A resistance-frame is frequently fitted with an electric device, by which it automatically adjusts the potential of the dynamo; when the E.M.F. drops, the apparatus moves the lever, or its equivalent, in such a way as to automatically short-cin suit certain of the resistance-springs, thereby bringing the pressure up again. Such an appliance is called an automatic dynamo-regulator or resistance.