PID (FB) ¶ FUNCTION_BLOCK PID Represents a PID controller Note The PID controller itself measures the elapsed time between two calls, however with a maximum accuracy of milliseconds. This might lead to rough running in case of short cycle times: For example in case of a cycle time of 1ms the PID sometimes might measure 2 ms, sometimes 0 ms. So if possible, for such cases better use PID_FIXCYCLE , where the cycle time can be set precisely. See PID_FIXCYCLE Note Consider that the controller parameters only get applied when used in the manual mode at a start, a reset or at a change down. A PID controller continuously calculates an error value e(t) as the difference between a desired set point and a measured process variable. The PID controller applies a correction based on proportional, integral, and derivative terms (sometimes denoted P, I, and D respectively) which give their name to the controller type. P accounts for present values of the error. For example, if the error is large and positive, the control output will also be large and positive. I accounts for past values of the error. For example, if the current output is not sufficiently strong, the integral of the error will accumulate over time, and the controller will respond by applying a stronger action. D accounts for possible future trends of the error, based on its current rate of change.[1] As a PID controller relies only on the measured process variable, not on knowledge of the underlying process, it is broadly applicable. By tuning the three parameters of the model, a PID controller can deal with specific process requirements. The response of the controller can be described in terms of its responsiveness to an error, of the degree to which the system overshoots a setpoint, and of the degree of any system oscillation. The use of the PID algorithm does not guarantee optimal control of the system or even its stability. Y_OFFSET , Y_MIN and Y_MAX serve for transformation of the manipulated variable within a prescribed range. MANUAL can be used to switch to manual operation; RESET can be used to re-initialize the controller. In normal operation ( MANUAL = RESET = LIMITS_ACTIVE = FALSE ) the controller calculates the controller error e as difference from SET_POINT – ACTUAL , generates the derivation with respect to time \(\frac{\delta e}{\delta t}\) and stores these values internally. The output Y is the manipulated variable unlike the PD controller contains an additional integral part, and is calculated as follows: \(Y = KP \cdot (e + \frac{1}{TN} \int e dt + TV \frac{\delta e}{\delta t}) + Y_{OFFSET}\) So besides the P-part also the current change of the controller error (D-part) and the history of the controller error (I-part) influence the manipulated variable. The PID controller can be easily converted to a PI-controller by setting TV=0 . Because of the additional integral part, an overflow can come about by incorrect parameterization of the controller, if the integral of the error e becomes to great. Therefore for the sake of safety a BOOLean output called OVERFLOW is present, which in this case would have the value TRUE. This only will happen if the control system is instable due to incorrect parameterization. At the same time, the controller will be suspended and will only be activated again by re-initialization. Note As long as the limitation for the manipulated variable ( Y_MIN , Y_MAX ) is active, the integral part will be adapted, like if the history of the input values had automatically effected the limited output value. If this behaviour is not wanted, the following workaround is possible: Switch off the limitation at the PID controller ( Y_MIN>=Y_MAX ) and instead apply the LIMIT operator (IEC standard) on output value Y (see an example in the figure below). Note It is not necessary to readjust the controller parameters (KP, TN, TV) if the cycle time changes. Temperature control with PID and LIMIT ¶ See in the following figure a simple example of using the PID module for temperature control and in combination with the LIMIT operator. The input of the actual temperature is simulated by giving a constant value via ActualTemperature. InOut: Scope Name Type Initial Comment Input ACTUAL REAL Current value, process variable SET_POINT REAL Desired value, set point KP REAL Proportionality const. P TN REAL Reset time I [sec] TV REAL Rate time, derivative time D [sec]. If set to 0, then it works as PI controller Y_MANUAL REAL Y is set to this value as long as MANUAL = TRUE Y_OFFSET REAL Offset for manipulated variable Y_MIN REAL Minimum value for manipulated variable Y_MAX REAL Maximum value for manipulated variable MANUAL BOOL TRUE : Manual: Y is not influenced by controller FALSE : Controller determines Y RESET BOOL TRUE : Set Y output to Y_OFFSET and reset integral part Output Y REAL Manipulated variable, set value LIMITS_ACTIVE BOOL FALSE TRUE : Y has exceeded the given limits Y_MIN , Y_MAX and is limited to these values OVERFLOW BOOL FALSE Overflow in integral part

PID_FIXCYCLE (FB) ¶ FUNCTION_BLOCK PID_FIXCYCLE Represents a PID controller for which the cycle time can be set manually A PID controller continuously calculates an error value e(t) as the difference between a desired set point and a measured process variable. The PID contoller applies a correction based on proportional, integral, and derivative terms (sometimes denoted P, I, and D respectively) which give their name to the controller type. Note For fast and fix tasks it is recommended to use PID_FIXCYCLE instead of PID, because the cycle time is defined accurately, whereas PID only can measure the cycle time with a maximum accuracy of milliseconds. In case of very short cycles (1ms) this might lead to a rough run (see PID). Note It is not necessary to readjust the controller parameters (KP, TN, TV) if the cycle time changes. For more information see: PID . InOut: Scope Name Type Initial Comment Input ACTUAL REAL Actual value, process variable SET_POINT REAL Desired value, set point KP REAL Proportionality const. P TN REAL Reset time I in sec TV REAL Rate time, derivative time D in sec Y_MANUAL REAL Y is set to this value as long as MANUAL = TRUE Y_OFFSET REAL Offset for manipulated variable Y_MIN REAL Minimum value for manipulated variable Y_MAX REAL Maximum value for manipulated variable MANUAL BOOL TRUE : Manual : Y is not influenced by controller FALSE : Controller determines Y RESET BOOL TRUE : Sets Y output to Y_OFFSET and reset integral part CYCLE REAL Time in s between two calls Output Y REAL Manipulated variable, set value LIMITS_ACTIVE BOOL FALSE TRUE : Set value would exceed limits Y_MIN , Y_MAX OVERFLOW BOOL FALSE Overflow in integral part

Datatypes ¶ GEN_MODE (Enum) POINT (Struct) PT_SIZE (Struct)

GEN_MODE (ENUM) ¶ TYPE GEN_MODE : InOut: Name Initial TRIANGLE 0 TRIANGLE_POS 1 SAWTOOTH_RISE 2 SAWTOOTH_FALL 3 RECTANGLE 4 SINUS 5 COSINUS 6 SINE 7 COSINE 8

ROTATION_DIFFERENCE (FB) ¶ FUNCTION_BLOCK ROTATION_DIFFERENCE Computes the difference between two UINT values without rollover This function block receives two values e.g. from an motion encoder and calculates the difference between the two positions. If curValue := 65535 and lastValue := 15, a negative rotation is detected and diffValues := -16. If lastValue := 65535 and curValue := 15, a positive rotation is detected and diffValues := 16. InOut: Scope Name Type Comment Input curValue UINT Current value of rotation lastValue UINT Last value of rotation Output diffValues INT Difference between the two values without rollover

STATISTICS_INT (FB) ¶ FUNCTION_BLOCK STATISTICS_INT Calculates minimum, maximum, and average of an input value The calculation is done over time and can be reset in order to start a fresh caluclation/statistic. InOut: Scope Name Type Initial Comment Input IN INT Input value RESET BOOL TRUE : AVG set to 0, MN and MX to minimum and maximum possible integer values Output MN INT 32767 Minimum value MX INT -32768 Maximum value AVG INT Average value

STATISTICS_REAL (FB) ¶ FUNCTION_BLOCK STATISTICS_REAL Calculates minimum, maximum, and average of an input value The calculation is done over time and can be reset in order to begin a fresh caluclation/statistic. InOut: Scope Name Type Initial Comment Input IN REAL Input value RESET BOOL TRUE : AVG set to 0, MN and MX to minimum and maximum possible real values Output MN REAL 3.402823466E+38 Minimum value MX REAL -3.402823466E+38 Maximum value AVG REAL Average value

VARIANCE (FB) ¶ FUNCTION_BLOCK VARIANCE Calculates the mathematical variance of a variable over time In this case, over time means that the variance is extended for each call to the function block until a reset is done. InOut: Scope Name Type Comment Input IN REAL Input value RESET BOOL Reset Output OUT REAL Variance

Signals ¶ BLINK (FunctionBlock) FREQ_MEASURE (FunctionBlock) GEN (FunctionBlock)

BLINK (FB) ¶ FUNCTION_BLOCK BLINK Simulates a blinking signal (turning on and off for specific durations) InOut: Scope Name Type Comment Input ENABLE BOOL TRUE : Starts blinking FALSE : Stops blinking whereas OUT keeps its value TIMELOW TIME Time for OUT to be FALSE TIMEHIGH TIME Time for OUT to be TRUE Output OUT BOOL Output value, starts with FALSE and switches between TRUE and FALSE for the given high and low times