Non-signal switches requirements
The category of non-signal switches includes all varieties of switch that are placed in supply voltage lines, whether DC or AC, or which make and break contacts between DC levels that are not of critical amplitudes. Switches that are used in power supply leads will often have a mechanical action that provides for a rapid switchover, the snap-over type of mechanism that is illustrated in Figure ll.l. This provides a form of mechanical hysteresis, so that the position of the switch operating mechanism for making connection is not the same as the position for breaking connection. This snap-over action is not provided on all types of mechanical switches, however, and may be undesirable if a slow opening rate is required. In this chapter we shall confine our attention to manually operated switches, because any discussion of relays and contactors would extend the scope of the book too far into more specialized topics.
The simplest switch requirement to fulfil is for high-voltage (in the region of a few hundred volts), low-current, non-inductive load, AC or DC. For such a requirement, the type of switch and the contact materials are not particularly critical and the only stipulation is that the switch should be adequately rated for the voltage and current required. The value of contact resistance, for example, will make only a small difference in the voltage across the closed contacts, and this difference will be negligible compared to the voltage that the switch is handling. The non-reactive load implies that there will be no current surge on making connection, nor any voltage surge on breaking connection. Because of this, no special precautions have to be taken against arcing, and a long life can be expected. Only if special environmental conditions have to be allowed for will there be any difficulty in selecting a suitable switch from any supplier.
Rather more care needs to be taken if the switching has to cope with high currents at high voltage, and considerably more care is needed if surges exist at switch-on or at switch-off. Taking these one at a time, the handling of high currents requires large contact areas, and contact materials that are of low contact resistance. The dissipation of power at the contact is equal to I2R, where I is the current (steady or surge) and R is the value of contact resistance. For example, for a current of l0 A and a contact resis- tance of 5 mD the power dissipated at the contact is 0.005 x l00 = 0.5 V. This may seem insignificant, but a lot depends on how large the contact area is, and how well the contacts can dissipate the resulting rise of tempera- ture. Just to put these figures into perspective, the 5 inch, lkV hotplate on an electric cooker will typically have a total surface area of l2 667 mm2, giving a dissipation of 0.07 Vjmm2. To achieve the same dissipation per unit area for switch contacts dissipating 0.5 V would require an area of contact of about 7 mm2, corresponding to a good fit of contacts of diameter 3 mm. From all this, you can see that a contact resistance of only 5 mD for a l0 A switch is by no means negligibly low, and that switch contacts, if thermally isolated, would run very hot.
Even if the voltage drop across the contact resistance is negligible, the effect of heat dissipation may not be, so that the size of the contact resistance and the area of the contacts are important considerations in any switch that handles large currents in the range of lA upwards. These factors will, of course, have been allowed for in the rating of the switch, but not all man- ufacturers will calculate their ratings making the same assumptions about thermal dissipation and the ratio of true contact area to measured area. If a large-current switch is intended for a sensitive application, in which failure would entail considerable consequent damage (for example, shutting down a complete installation), then it would be wise either to use more generously rated switches, or to measure the temperature of a few samples after an 8-hour period at full rated current in the expected maximum ambient temperature. The contact resistance should also be measured before and after such a test, and any significant increase will point to local overheating, which will not necessarily show up in the tem- perature measurements.
Switching for low-voltage supplies brings another set of problems, depending on whether the current is large or small. Most low-voltage switching is likely to involve large currents, so that a low contact resistance is an important requirement from both the dissipation and the voltage- drop considerations. In addition, low-voltage DC supplies are likely to use large capacitors on each side of the switch, and the capacitor on the load side will charge at the instant of switch-on, causing a surge current that can be very large (at least l0 times the average supply current). If this might cause difficulties, non-linear surge-suppressing resistors can be included in the supply line, but the voltage drop across such a device is usually prohibitive for low-voltage supplies. If possible, circuits should be arranged so that no large capacitors are connected on the load side of the switch between supply voltage lines. Unfortunately, the requirement for the minimum value of contact resistance in a switch for low-voltage, high- current applications often conflicts with the equally important requirement of long life. As always, there is a trade-off between contact resistance and contact durability.
The easiest requirement to fulfil for low-voltage supplies is for low current (in the range of a few mA to a few hundred mA). At these currents, contact resistance is not the main consideration, small contact areas can be used, and only the possibility of surge current or voltage can make for any difficulty in switch selection. The mechanical action of the switch is seldom important. There is no need for a snap action, so the full range of actions can be considered, and the physical size of the switch is unimportant from the electrical point of view. Switch size should, however, be considered from the ergonomic point of view. Many very small switches are available, but they are decidedly difficult to use simply because they are so small. Long lines of such switches should be avoided, no matter how satisfying they may look, because it is highly likely that at some stage a user will unintentionally switch over two at once. Vhen switch operating levers are so close that a human fingertip can cover more than one, trouble can be expected eventually.