PID overview: About closed loop control

The underlying principle of PID control is feedback or closed loop control. PID is only one of many possible forms of feedback control. PID is not always the best form of control. You need to have a good understanding of the process you wish to control before you can decide if PID control is the appropriate choice.

The Process is some apparatus that produces an Output that can be varied by changing an Control Input, which is also referred to as a forcing function. The output is continuously variable, like the light from a bulb controlled by a dimmer, as distinct from discontinuous like the light from a bulb controlled by an on/off switch. The input to the process need not be continuously adjustable, but usually is. Notice that in the case of a light bulb example the process consists of the dimmer or switch plus the bulb, and its output is the light produced. (To be more precise, the process is the conversion of electricity into light. The light bulb and dimmer/switch are the apparatus that performs the process.)

The Controller is a device (in our case a SPLat controller) that measures the output of the process, compares it with the desired output (called the Setpoint) and then applies a corrected Control Input (the forcing function) to the process.

This setup is called feedback control because (a measure of) the process output is fed back into the controller and hence (after some number crunching) back into the process. It is also called closed loop control, because the process and the controller are connected together "head to tail" into a closed loop.

Examples of closed loop control:

• A room thermostat (which by the way is normally not a PID controller).
• A motor car cruise control.

• You and your car, as you maintain a constant speed on the freeway. YOU are the controller in this case, you might like to keep this in mind as you gain understanding of the mechanisms behind automatic control.

• A petroleum refinery (which actually has hundreds of interconnected control loops)

Notice that the output of the process (or at least a measure of it) becomes an input to the controller, and the output of the controller becomes the control input to the process.

The controller has a second input, the setpoint, also referred to as a Target Value. The setpoint is what tells the controller the desired value of the process output.

The process usually has one or more additional inputs. These are things like electric power, chemical feedstock etc. They do not directly control the output but they can certainly influence it. For example, if the supply voltage drops the light output of the bulb will drop. Often, too, the amount of loading on the output may affect the level of the output. For example, a sudden head wind will slow down your car and result in a higher throttle setting being needed to maintain the same speed.

The purpose of closed loop control is to maintain the process output at the setpoint irrespective of external influences on the process such as mains voltage, loading, ambient temperature etc.