Description

Note These SFB's only exist in a few CPU's. Please consult the manual for your specific CPU to determine whether the SFB's required for integrated control facilities are provided by your CPU.

Introduction

SFB 41 is employed to control technical processes of automation systems with continuous input and output quantities. Individual functions of the PID-controller can be enabled or disables by means of parameters to adjust the controller to the requirements of the control process.

Application

The controller can be used as a PID set-point controller, as an independent control unit or in multi-loop control configurations as cascaded, combined or proportional controller. Its operation is based on the PID control algorithm of the sampling controller with analog output signals. This can be extended by a pulse shaper to generate pulse-width modulated output signals that are required for on-off controls or three-mode controls of proportional actuators.

Description

In addition to the functions for the set-point and the actual value branches, SFB 41 also includes a complete PID-controller with continuous output of control-variables as well as manual adjustment options for the manipulated value.
The following description contains an explanation of the different functions:

Set-point branch
The set-point is defined as a floating-point formatted value at input SP_INT.

Actual value branch
The actual value can be supplied in peripheral or in floating point format.
Peripheral value PV_PFC is converted to a floating point number of -100.... +100 % by the function CRP_IN in accordance with the following formula:

100
output of CRP_IN = PV_PFC * ¾ ¾ ¾ ¾
27648

The function PV_NORM can normalize the output of CRP_IN in accordance with the following formula:

output of PV_NORM = (output of CRP_IN) * PV_FAC + PV_OFF

where PV_FAC is preset to 1 and PV_OFF to 0.

Calculation of difference (error) value
The error results from the difference between set-point value and actual value. It is possible that a small amount of oscillation occurs due to the quantization of the controlling variables (e.g. PULSEGEN when pulse-width modulation is used). To suppress this oscillation the difference value is routed via a dead band (DEADBAND). The dead band is deactivated when input parameter DEADB_W is set to 0.

PID-Algorithm
The PID-algorithm operates as position controller algorithm, where the proportional, integral (INT) and the differential portion (DIF) are paralleled and may be included or excluded individually. This can be used to implement P, PI, PD and PID controllers as well as pure I controllers.

Manual control
You can select between manual or automatic mode. In manual mode the control variable tracks the manual value.
Internally the integrator (INT) is set to LMN - LMN_P - DISV and the differentiator (DIF) is set to 0. These are then synchronized internally to provide for a smooth changeover into automatic mode.

Control value processing
The LMNLIMIT function limits the manipulated values to a range of values that can be preset. Message bits provide an indication when the input values exceed the limits.
The output of LMNLIMIT can be normalized by the function LMN_NORM in accordance with the following formula:

LMN = (output of LMNLIMIT) * LMN_FAC + LMN_OFF

were LMN_FAC is preset to 1 and LMN_OFF to 0.

Since the manipulated value is also available in peripheral format, floating point value LMN is converted to a peripheral value by means of the CRP_OUT function:

27648
LMN_PFC = LMN * ¾ ¾ ¾ ¾
100

Addition of a disturbance quantity
It is possible to add a disturbance quantity or value via input DISV.