The Quick Switch Guide

The system by which switches are specified is, to say the least, not simple. It's practical, and now so common that it is likely to endure for a good while, but it can be very challenging for those just learning the ropes.

There are a few key factors that go into a switch's designation, and we're going to more or less cover those here.

Poles and throws

The primary method used to specify a switch is how many poles it has, and how many throws it has. If those terms make no sense to you, you're not alone.

The basic image of a switch in most people's mind is a piece of wire connected to a circuit on one side, and then free to be connected or not on its other end to more circuitry. This is the simplest switch that can be constructed, and sees applications all over the place. It is known as an SPST switch, where SPST stands for Single Pole, Single Throw. It has only two possible states: on, and off.

We could also put two of these together, next to each other. We could control them with a single mechanism so they are controlled together but are not electrically connected. This would be a DPST switch; Double Pole, Single Throw. It has two separate poles, each of which can be connected (or not) to one contact. Again, there are two possible states: on and off. But the state of one pole is duplicated for the other.

But what if we take our SPST switch and give it a new option? Instead of just being connected (or not) to one point in the circuit, what if we gave it two possible places to connect? This would be an SPDT switch; Single Pole, Double Throw. It can take two different states (or perhaps three, if the switch is built to allow a state where the switch doesn't connect to either point).

Multi-throw switches are built so that one set of pins is always connected to the circuit, and then wires running from those pins can be connected to other pins. The pins that are always connected are sometimes called common pins, because they are always connected regardless of the state the switch is in. When you change the state of the switch, you are changing what gets connected to the common pins. Very often common pins are situated in the middle of the switch to make them easier to locate on switches where markings may have become difficult to read.

From these basic concepts, we can build all manner of switches. We can take our SPDT switch and combine 2 to make a DPDT (Double Pole, Double Throw) switch. Such a switch would have two common pins, and connect them to one of two sets of other pins depending on its state.

We can go bigger too. We can make a 3PST switch that has 3 separate poles, for instance. By tradition, everything above 2 is denoted with a number, not a letter. We could build a 24PST switch if needed. Or SP24T. Or for that matter, 24P24T. As you may have guessed, bigger and more complex switches cost more. They also may not be as widely available as simpler configurations.


All switches fall into one of two categories: persistent or momentary.

Persistent or latching switches, after being put into a particular state, stay in that state until made to do otherwise. They are useful in many applications in which a user wants some event to happen continuously. Most light switches, for instance, are persistent.

Momentary switches stay in a state only so long as they are forced to. When the controlling input is removed, they revert to a different state. They are useful in many applications in which a user wants some event to happen only for a limited amount of time. Car window controls and keyboard key switches are momentary.

It is possible (and sometimes desirable) to combine these properties in a single switch. Some SPST switches, for instance, have an "off" state where they are disconnected and then two "on" states in which they are connected. One of these "on" states is persistent; the other is momentary.

Normal states

Momentary switches have a normal or default state. This is the state they return to when not supplied with an input.

Normally open (NO) switches default to a disconnected state. The term "open" refers to the fact that this state is an open circuit.

Normally closed (NC) switches default to a connected state. The term "closed" refers to the fact that this state is a closed circuit.

Control types

There are lots of ways to control a switch. Generally this refers to the physical mechanism that is used to move a switch from one state to another. it might be a lever, or a button, or even an electromagnet (see below).

Pushbutton switches are, as you might guess, controlled by a button. This generally means something you push down (toward the body of the switch). Computer power buttons, keyboard keys, and the buttons on most television remotes are all pushbutton switches.

Toggle switches have a lever arm that sticks out and can be flipped between states. Older light switches are toggle switches.

Rocker switches are similar to toggle switches, but instead of one big lever they have a broader, nearly flat surface. Very often, they have built-in springs that pull them back to a particular state after they are pressed (making them momentary switches). Newer light switches are often rocker switches, as are the switches that control car door locks.

Slide switches have an assembly that slides from one point to another. They are usually built with relatively few throws, but may have a large number of poles. Slide switches are usually not momentary. Many small electronics have slide switches to control their power.

DIP switches are very small slide switches. They are built for compactness, not durability, and are usually used inside devices. DIP switches are often used in devices to set configuration information. DIP switches are usually persistent SPST switches, and are very often built into packages that contain multiple, individually controllable switches. Many remote controls contain DIP switches.

Rotary switches have an element that rotates and, in so doing, connects to different sets of contacts. Rotary switches are often used for complex configurations, because they allow a great deal of flexibility and can accomodate large numbers of poles and/or throws. Many ceiling fans are controlled by rotary switches.

Snap-action switches have a long, loosely-attached metal arm that "snaps". The lever usually depresses a sort of pushbutton switch. Some have an element that rotates instead of an arm; the rotating element has a spring that pulls it to some state if no force is applied. Snap-action switches are almost always momentary. They are useful in applications that want a "hair trigger". Snap-action switches are often used in intrusion detectors.

Magnetic switches function by detecting the presence (or absence) of a magnetic field. Reed switches are a major group; they are activated by the presense of a magnetic field of sufficient strength. They are very similar to relays, with the control coil replace by some external magnet. Hall effect sensors are sometimes labeled as switches; they report the presence and intensity of magnetic fields. Magnetic switches are often used in home alarm systems to detect whether doors and windows are closed.

Specialty switches of all sorts exist. They may detect heat (some home dimmer switches), variations in capacitance and inductance (some elevator call buttons), or virtually any particular phenomenon.

What about relays?

So we've talked all about different kinds of switches and such, and every now and again I've made reference to relays. But just what are relays?

Relays are electromagnetically controlled switches. A basic relay has a coil on one side and a moveable arm on the other. When the coil is energized, it produces a magnetic field that moves the arm and makes or breaks contact. It is very similar to a reed switch, but with a built-in electromagnet.

Relays can be constructed in all sorts of configurations - SPST, DPDT, momentary, latching, etc. - just like switches. The only substantive difference is that they aren't controlled by a moving lever, or knob, or what have you, but by a control voltage.

  • RSS feed StumbleUpon Digg Yahoo! My Web 2.0