CH 8

Chapter
PID Loop Operation

8
In This Chapter...
D4-454 PID Control.................................................................................................... 8-2
Introduction to PID Control....................................................................................... 8-4
Introducing D4-454 PID Control................................................................................ 8-6
PID Loop Operation.................................................................................................... 8-9
Ten Steps to Successful Process Control.................................................................. 8-16
PID Loop Setup......................................................................................................... 8-18
PID Loop Tuning....................................................................................................... 8-42
Using the Special PID Features................................................................................ 8-54
Ramp/Soak Generator.............................................................................................. 8-60
DirectSOFT Ramp/Soak Example............................................................................. 8-65
Cascade Control........................................................................................................ 8-67
Time-Proportioning Control..................................................................................... 8-70
Feedforward Control................................................................................................ 8-72
PID Example Program.............................................................................................. 8-74
Troubleshooting Tips................................................................................................ 8-77
Glossary of PID Loop Terminology.......................................................................... 8-79
Bibliography ............................................................................................................ 8-81
Chapter 8: PID Loop Operation
D4-454 PID Control

D4-454 PID Control Features
Along with control functions discussed in this manual, the D4-454 PLC features PID process
control capability. The D4-454 PID process control loops offer the same features offered in
much larger PLCs. The primary features are:
• Up to 16 PID loops, individual programmable sample rates
• Manual, Automatic and Cascade loop operation modes
• Two types of bumpless transfer available
• Full-featured alarms
• Ramp/soak generator with up to 16 segments
• Auto Tuning
The D4-454 CPU has process control loop capability in addition to ladder program execution.
You can select and configure up to sixteen loops. All sensor and actuator wiring connects
directly to DL405 analog modules. All process variables, gain values, alarm levels, etc.,
associated with each loop reside in a Loop Variable Table in the CPU. The D4-454 CPU reads
process variable (PV) inputs during each scan. Then, it makes PID loop calculations during a
dedicated time slice on each PLC scan, updating the control output value. The control loops
use a Proportional-Integral-Derivative (PID) algorithm to generate the control output. This
chapter describes how the loops operate, and how to configure and tune the loops.
Analog Input
D4-454
PID Loop Calculations
ations
Manufacturing Process
Analog Output
DirectSOFT programming software, release 6.1, or later, is used for configuring analog control
loops in the D4-454. DirectSOFT uses dialog boxes to help you set up the individual loops.
After completing the setup, you can use DirectSOFT’s PID Trend View to tune each loop. The
configuration and tuning selections you make are stored in the D4-454’s V-memory , which
can be set as retentive. The loop parameters also may be saved to disk for recall later.
8-2 D4-454 PLC User Manual, 1st Edition, Rev. B

PID Loop Feature Specifications

Number of loops Selectable, 16 maximum
CPU V-memory needed 32 words (V locations) per loop selected, 64 words if using ramp/soak
PID algorithm Position or Velocity form of the PID equation
Control Output polarity Selectable direct-acting or reverse-acting
Error term curves Selectable as linear, square root of error, and error squared
Loop update rate (time 0.05 to 99.99 seconds, user programmable
between PID calculation)
0.05 seconds for 1 to 4 loops
Minimum loop update rate 0.1 seconds for 5 to 8 loops
0.2 seconds for 9 to 16 loops
Loop modes Automatic, Manual (operator control), or Cascade control
Ramp/Soak Generator Up to 8 ramp/soak steps (16 segments) per loop with indication of ramp/soak step number
PV curves Select standard linear, or square-root extract (for flow meter input)
Set Point Limits Specify minimum and maximum setpoint values
Process Variable Limits Specify minimum and maximum Process Variable values
Proportional Gain Specify gains of 0.0 to 99.99
Integrator (Reset) Specify reset time of 0.0 to 99.99 in units of seconds or minutes
Derivative (Rate) Specify the derivative time from 0.0 to 99.99 seconds
Rate Limits Specify derivative gain limiting from 1 to 20
Bumpless Transfer I Automatically sets the bias equal to the control output and the setpoint equal to the process
variable when control switches from manual to automatic.
Bumpless Transfer II Automatically sets the bias equal to the control output when control switches from manual
to automatic.
Step Bias Provides proportional bias adjustment for large setpoint changes
Anti-windup (Freeze Bias) For position form of PID, this inhibits integrator action when the control output reaches 0%
or 100 % (speeds up loop recovery when output recovers from saturation).
Error Deadband Specify a tolerance (plus and minus) for the error term (SP–PV), so that no change in
control output value is made.
Alarm Feature Specifications

PV Alarm Hysteresis Specify 1 to 200 (word/binary) does not affect all alarms, such as PV Rate-of-Change Alarm
PV Alarm Points Select PV alarm settings for Low–low, Low, High, and High-high conditions
PV Deviation Specify alarms for two ranges of PV deviation from the setpoint value
Rate of Change Detect when PV exceeds a rate of change limit you specify
D4-454 PLC User Manual, 1st Edition, Rev. B 8-3

Introduction to PID Control

What is PID Control?
In this discussion, we will explain why PID control is used in process control instead of trying
to provide control by simply using an analog input and a discrete output.
There are many types of analog controllers available, and the proper selection will depend upon
the particular application. There are two types of analog controllers that are used throughout
industry:
1. The ON-OFF controller, sometimes referred to as an open loop controller.
2. The PID controller, sometimes called a closed loop controller.
Regardless of type, analog controllers require input signals from electronic sensors such as
pressure, differential pressure, level, flow meter or thermocouples. As an example, one of the
most common analog control applications is located in your house for controlling either heat
or air conditioning, the thermostat.
You wish for your house to be at a comfortable temperature so you set a thermostat to a desired
temperature (setpoint). You then select the comfort mode, either heat or A/C. A temperature
sensing device, normally a thermistor, is located within the thermostat. If the thermostat is set
for heat and the setpoint is set for 69!, the furnace will be turned on to provide heat at, normally,
2! below the setpoint. In this case, it would turn on at 67!. When the temperature reaches 71!,
2! above setpoint, the furnace will turn off. In the opposite example, if the thermostat is set
for A/C (cooling), the thermostat will turn the A/C unit on/off opposite the heat setting. For
instance, if the thermostat is set to cool at 76!, the A/C unit will turn on when the sensed
temperature reaches 2! above the setpoint or 78!, and turn off when the temperature reaches
74!. This would be considered to be an ON-OFF controller. The waveform below shows the
action of the heating cycle. Note that there is a slight overshoot at the turn-off point, also a
slight undershoot at the turn-on point.
71° OFF OFF
69° SETPOINT
67°
ON ON ON
TIME
The ON-OFF controller is used in some industrial control applications, but is not practical in
the majority of industrial control processes.
The most common process controller that is used in industry is the PID controller.

The PID controller controls a continuous feedback loop that keeps the process output (control
variable) flowing normally by taking corrective action whenever there is a deviation from the
desired value (setpoint) of the process variable (PV) such as, rate of flow, temperature, voltage,
etc. An error occurs when an operator manually changes the setpoint or when an event (valve
opened, closed, etc.) or a disturbance (cold water, wind, etc.) changes the load, thus causing a
change in the process variable.
The PID controller receives signals from sensors and computes corrective action to the actuator
from a computation based on the error (Proportional), the sum of all previous errors (Integral)
and the rate of change of the error (Derivative).
We can look at the PID controller in more simple terms. Take the cruise control on an
automobile as an example. Let’s say that we are cruising on an interstate highway in a car
equipped with cruise control. The driver decides to engage the cruise control by turning it
ON, then he manually brings the car to the desired cruising speed, say 70 miles per hour. Once
the cruise speed is reached, the SET button is pushed fixing the speed at 70mph, the setpoint.
Now, the car is cruising at a steady 70mph until it comes to a hill to go up. As the car goes up
the hill, it tends to slow down. The speed sensor senses this and causes the throttle to increase
the fuel to the engine. The vehicle speeds up to maintain 70mph without jerking the car and
it reaches the top at the set speed. When the car levels out after reaching the top of the hill it
will speed up. The speed sensor senses this and signals the throttle to provide less fuel to the
engine, thus, the engine slows down allowing the car to maintain the 70mph speed. How does
this application apply to PID control? Lets look at the function of P, I and D terms:
• Proportional - is commonly referred to as Proportional Gain. The proportional term is the corrective
action which is proportional to the error, that is, the change of the manipulated variable is equal to
the proportional gain multiplied by the error (the activating signal). In mathematical terms:
Proportional action = proportional gain X error
Error = Setpoint (SP) - Process Variable (PV)
• Applying this to the cruise control, the speed was set at 70mph which is the Setpoint. The speed
sensor senses the actual speed of the car and sends this signal to the cruise controller as the Process
Variable (PV). When the car is on a level highway, the speed is maintained at 70mph, thus, no error
since the error would be SP - PV = 0. When the car goes up the hill, the speed sensor detected a
slow down of the car, SP-PV = error. The proportional gain would cause the output of the speed
controller to bring the car back to the setpoint of 70mph. This would be the Controlled Output.
• Integral - this term is often referred to as Reset action. It provides additional compensation to the
control output, which causes a change in proportion to the value of the error over a period of time.
In other words, the reset term is the integral sum of the error values over a period of time.
• Derivative - this term is referred to as rate. The Rate action adds compensation to the control
output, which causes a change in proportion to the rate of change of error. Its job is to anticipate the
probable growth of the error and generate a contribution to the output in advance.

Introducing D4-454 PID Control

The D4-454 is capable of controlling a process variable such as those already mentioned. As
previously mentioned, the control of a variable, such as temperature, at a given level (setpoint)
as long as there are no disturbances (cold water) in the process.
The D4-454 PLC has the ability to directly accept signals from electronic sensors, such as
thermocouples, pressure transducers, VFDs, etc. These signals may be used in mathematically
derived control systems.
In addition, the D4-454 has built-in PID control algorithms that can be implemented. The
basic function of PID closed loop process control is to maintain certain process characteristics
at desired setpoints. As a rule, the process deviates from the desired setpoint reference as a
result of load material changes and interaction with other processes. During this control,
the actual condition of the process characteristics (liquid level, temperature, pressure, motor
control, etc.) is measured as a process variable (PV) and compared with the target setpoint (SP).
When deviations occur, an error is generated by the difference between the process variable
(actual value) and the setpoint (desired value). Once an error is detected, the function of the
control loop is to modify the control output in order to force the error to zero.
The DL405 PID control provides feedback loops using the PID algorithm. The control
output is computed from the measured process variable as follows:
Let:
Kc = proportional gain
Ti = Reset or integral time
Td = Derivative time or rate
SP = Setpoint
PV(t) = Process Variable at time “t”
e(t) = SP-PV(t) = PV deviation from setpoint at time “t” or PV error.
Then:
M(t) = Control output at time “t”
t
M(t) = Kc [ e(t) + 1/Ti * e(x) dx + Td d/dt e(t) ] + M
0 o
The analog input module receives the process variable in analog form along with an operator
entered setpoint; the CPU computes the error. The error is used in the algorithm computation
to provide corrective action at the control output. The function of the control action is based
on an output control, which is proportional to the instantaneous error value. The integral
control action (reset action) provides additional compensation to the control output, which
causes a change in proportion to the value of the change of error over a period of time. The
derivative control action (rate change) adds compensation to the control output, which causes
a change in proportion to the rate of change of error. These three modes are used to provide
the desired control action in Proportional (P), Proportional-Integral (PI), or Proportional-
Integral-Derivative (PID) control fashion.

Standard DL405 analog input modules are used to interface to field transmitters to obtain the
PV. These transmitters normally provide a 4–20 mA current or an analog voltage of various
ranges for the control loop.
For temperature control, thermocouple or RTD can be connected directly to the appropriate
module. The PID control algorithm, residing in the CPU memory, receives information from
the user program, primarily control parameters and setpoints. Once the CPU makes the PID
calculation, the result may be used to directly control an actuator connected to a 4-20mA
current output module to control a valve.
With DirectSOFT, additional ladder logic programming, both time proportioning (e.g.,
heaters for temperature control) and position actuator (e.g., reversible motor on a valve) type
of control schemes can be easily implemented. This chapter will explain how to set up the PID
control loop, how to implement the software and how to tune the loop.
The following block diagram shows the key parts of a PID control loop. The path from the
PLC to the manufacturing process and back to the PLC is the closed loop control.
Loop Configuring External

and Monitoring Disturbances
PLC System
Setpoint Value Error Term Loop Control Output Manufacturing

+ Calculation Process
–
Process Variable

Process Control Definitions

Manufacturing Process – the set of actions that adds value to raw materials. The process can
involve physical changes and/or chemical changes to the material. The changes render the
material more useful for a particular purpose, ultimately used in a final product.
Process Variable – The controlled variable part of the process that you wish to control. It may
be temperature, pressure, level, flow, composition, density, the ratio of two streams, etc. Also
known as the actual value.
Setpoint – This is the target for the process variable. When all conditions of the process are
correct, the process variable will equal the setpoint.
Control Output – The result of the loop calculation, which becomes a command for the
process (such as the heater level in an oven). This is sometimes referred to as control variable.
Error Term – The algebraic difference between the process variable and the setpoint. This is
the control loop error, and is equal to zero when the process variable is equal to the setpoint
(desired) value. A well-behaved control loop is able to maintain a small error term magnitude.
Manipulated Variable – This is what is used to effect the controlled variable. For example, the
fuel used in a furnace might be manipulated in order to control the temperature.
Disturbance – Something in the system that changes such that corrective action is required.
For instance, when controlling a flow and the upstream pressure drops, the control valve must
open wider in order to keep flow constant. The drop in upstream pressure is the disturbance.
Final Control Element – This is the physical device used to control the manipulated variable.
Valves are probably the most widely used final control element.
Lag Time – The time it takes for the process to respond to a change in manipulated variable.
This is also known as the capacitance of the system. When you’re in the shower and you turn
up the hot water a little, the time it takes before the water gets hot is the lag time.
Dead Time – The time it takes for a change in the process to be recognized. Composition
analyzers and quality control are usually sources of significant dead time.
Loop Configuring – Operator-initiated selections which set up and optimize the performance
of a control loop. The loop calculation function uses the configuration parameters in real time
to adjust gains, offsets, etc.
Loop Monitoring – The function which allows an operator to observe the status and
performance of a control loop. This is used in conjunction with the loop configuring to
optimize the performance of a loop (minimize the error term).

PID Loop Operation

The Proportional–Integral–Derivative (PID) algorithm is widely used in process control. The
PID method of control adapts well to electronic solutions, whether implemented in analog
or digital (CPU) components. The D4-454 CPU implements the PID equations digitally by
solving the basic equations in software. I/O modules serve only to convert electronic signals
into digital form (or vice versa).
The D4-454 uses two types of PID controls: “position” and “velocity”. These terms usually
refer to motion control situations, but here we use them in a different sense:
• PID Position Algorithm – The control output is calculated so it responds to the displacement
(position) of the PV from the SP (error term).
• PID Velocity Algorithm – The control output is calculated to represent the rate of change (velocity)
for the PV to become equal to the SP.
Position Form of the PID Equation

Referring to the control output equation on page 8-6, the D4-454 CPU approximates the
output M(t) using a discrete position form of the PID equation.
Let:
Ts = Sample rate
Kc = Proportional gain
Ki = Kc * (Ts/Ti) = Coefficient of integral term
Kr = Kc * (Td/Ts) = Coefficient of derivative term
Ti = Reset or integral time
Td = Derivative time or rate
SP = Setpoint
PVn = Process variable at nth sample
en = SP – PVn = Error at nth sample
Mo = Value to which the controller output has been initialized
Then:
Mn = Control output at nth sample
n
Mn = Kc * en + Ki (ei + Kr (en - en-1) + Mo
i=1
This form of the PID equation is referred to as the position form since the actual actuator
position is computed. The velocity form of the PID equation computes the change in actuator
position. The CPU modifies the standard equation slightly to use the derivative of the process
variable instead of the error as follows:
n
Mn = Kc * en + Ki (ei + Kr (PVn - PVn-1) + Mo
i=1
These two forms are equivalent unless the setpoint is changed. In the original equation, a large
step change in the setpoint will cause a correspondingly large change in the error resulting in a
bump to the process due to derivative action. This bump is not present in the second form of
the equation.

The D4-454 also combines the integral sum and the initial output into a single term called the
bias (Mx). This results in the following set of equations:
Mxo = Mo
Mx = Ki * en + Mxn-1
Mn = Kc * en - Kr(PVn-PVn-1) + Mxn
The D4-454 by default will keep the normalized output M in the range of 0.0 to 1.0. This is
done by clamping M to the nearer of 0.0 or 1.0 whenever the calculated output falls outside
this range. The D4-454 also allows you to specify the minimum and maximum output limit
values (within the range 0 to 4095 in BCD if using 12 bit unipolar).
NOTE: The equations and algorithms, or parts of, in this chapter, are only for references. Analysis of these
equations can be found in most good text books about process control.
Reset Windup Protection

Reset windup can occur if reset action (integral term) is specified and the computation of the
bias term Mx is:
For example, assume the output is controlling a valve and the PV remains at some value
greater than the setpoint. The negative error (en) will cause the bias term (Mx) to constantly
decrease until the output M goes to 0 closing the valve. However, since the error term is still
negative, the bias will continue to decrease becoming ever more negative. When the PV finally
does come back down below the SP, the valve will stay closed until the error is positive for
long enough to cause the bias to become positive again. This will cause the process variable to
undershoot.
One way to solve the problem is to simply clamp the normalized bias between 0.0 and 1.0. The
D4-454 CPU does this. However, if this is the only thing that is done, then the output will not
move off 0.0 (thus opening the valve) until the PV has become less than the SP. This will also
cause the process variable to undershoot.
The D4-454 CPU is programmed to solve the overshoot problem by either freezing the bias
term, or by adjusting the bias term.

Freeze Bias
If the “Freeze Bias” option is selected when setting up the PID loop (discussed later) then the
CPU simply stops changing the bias (Mx) whenever the computed normalized output (M) goes
outside the interval 0.0 to 1.0.
M = Kc * en - Kr(PVn - PVn-1) + Mx
Mn = 0 if M < 0
Mn = M if 0 < M < 1
Mn = 1 if M > 1
Mxn = Mx if 0 < M < 1

Mxn = Mxn-1 otherwise
Thus in this example, the bias will probably not go all the way to zero so that, when the PV
does begin to come down, the loop will begin to open the valve sooner than it would have if
the bias had been allowed to go all the way to zero. This action has the effect of reducing the
amount of overshoot.
Adjusting the Bias

The normal action of the CPU is to adjust the bias term when the output goes out of range as
shown below.
M = Kc * en - Kr(PVn - PVn-1) + Mx
Mn = 0 if M < 0
Mn = M if 0 < M < 1
Mn = 1 if M > 1
Mxn = Mx if 0 < M < 1

Mxn = Mn - Kc * en - Kr(PVn - PVn-1) otherwise
By adjusting the bias, the valve will begin to open as soon as the PV begins to come down. If the loop is
properly tuned, overshoot can be eliminated entirely. If the output went out of range due to a setpoint
change, then the loop probably will oscillate because we must wait for the bias term to stabilize again.
The choice of whether to use the default loop action or to freeze the bias is dependent on the application.
If large, step changes to the setpoint are anticipated, then it is probably better to select the freeze bias
option (see page 8-34).

Step Bias Proportional to Step Change in SP

This feature reduces oscillation caused by a step change in setpoint when the adjusting bias
feature is used.
Mx = Mx * SPn / SPn-1 if the loop is direct acting
Mx = Mx * SPn-1 / SPn if the loop is reverse acting
Mxn = 0 if Mx < 0
Mxn = Mx if 0 < Mx < 1
Mxn = 1 if M > 1
Eliminating Proportional, Integral or Derivative Action

It is not always necessary to run a full three mode PID control loop. Most loops require only
the PI terms or just the P term. Parts of the PID equation may be eliminated by choosing
appropriate values for the gain (Kc), reset (Ti) and rate (Td) yielding a P, PI, PD, I and even
an ID and a D loop.
Eliminating Integral Action he effect of integral action on the output may be eliminated
T
by setting Ti = 9999. When this is done, the user may then
manually control the bias term (Mx) to eliminate any
steady-state offset.
Eliminating Derivative Action he effect of derivative action on the output may be
T
eliminated by setting Td = 0 (most loops do not require a
D parameter; it may make the loop unstable).
Eliminating Proportional Action A lthough rarely done, the effect of proportional term on
the output may be eliminated by setting Kc = 0. Since Kc
is also normally a multiplier of the integral coefficient (Ki)
and the derivative coefficient (Kr), the CPU makes the
computation of these values conditional on the value of
Kc as follows:
Ki = Kc * (Ts / Ti) if Kc Þ 0
Ki = Ts / Ti if Kc = 0 (I or ID only)
Kr = Kc * (Td / Ts) if Kc Þ 0
Kr = Td / Ts if Kc = 0 (ID or D only)
Velocity Form of the PID Equation

The standard position form of the PID equation computes the actual actuator position. An
alternative form of the PID equation computes the change in actuator position. This form
of the equation is referred to as the velocity PID equation and is obtained by subtracting the
equation at time “n” from the equation at time “n-1”.
The velocity equation is given by:
DMn = M - Mn-1
DMn = Kc * (en - en-1) + Ki * (PVn - 2 * PVn-1 + PVn-2)

Bumpless Transfer
The D4-454 loop controller provides for bumpless mode changes. A bumpless transfer from
manual mode to automatic mode is achieved by preventing the control output from changing
immediately after the mode change.
When a loop is switched from Manual mode to Automatic mode, the setpoint and Bias are
initialized as follows:
• Position PID Algorithm Velocity PID Algorithm
SP = PV SP = PV
Mx = M
The bumpless transfer feature of the D4-454 is available in two types: Bumpless I and Bumpless
II (see page 8-27). The transfer type is selected when the loop is set up.
Loop Alarms
The D4-454 allows the user to specify alarm conditions that are to be monitored for each loop.
Alarm conditions are reported to the CPU by setting up the alarms in DirectSOFT using the
PID setup alarm dialog when the loop is setup. The alarm features for each loop are:
• PV Limit – Specify up to four PV alarm points.
High-High
PV rises above the programmed High-High Alarm Limit.
High PV rises above the programmed High Alarm Limit.
Low PV fails below the Low Alarm Limit.
Low-Low PV fails below the Low-Low Limit.
• PV Deviation Alarm – Specify an alarm for High and Low PV deviation from the setpoint (Yellow
Deviation). An alarm for High High and Low Low PV deviation from the setpoint (Orange
Deviation) may also be specified. When the PV is further from the setpoint than the programmed
Yellow or Orange Deviation Limit the corresponding alarm bit is activated.
• Rate of Change – This alarm is set when the PV changes faster than a specified rate-of-change limit.
• PV Alarm Hysteresis – The PV Limit Alarms and PV Deviation Alarms are programmed
using threshold values. When the absolute value or deviation exceeds the threshold, the
alarm status becomes true. Real-world PV signals have some noise on them, which can
cause some fluctuation in the PV value in the CPU. As the PV value crosses an alarm
threshold, its fluctuations will cause the alarm to be intermittent and annoy process
operators. The solution is to use the PV Alarm Hysteresis feature.

Loop Operating Modes

The D4-454 loop controller operates in one of three modes, either Manual , Automatic or
Cascade.
Manual
In manual mode, the control output is determined by the operator, not the loop controller.
While in manual mode, the loop controller will still monitor all of the alarms including High-
High, High, Low, Low-Low, Yellow deviation, Orange deviation and Rate-of-Change.
Automatic
In automatic mode, the loop controller computes the control output based on the programmed
parameters stored in V-memory. All alarms are monitored while in automatic.
Cascade
Cascade mode is an option with the D4-454 PLC and is used in special control applications. If
the cascade feature is used, the loop will operate as it would if in automatic mode except for the
fact that a cascaded loop has a setpoint which is the control output from another loop.
Special Loop Calculations

Reverse Acting Loop
Although the PID algorithm is used in a direct, or forward, acting loop controller, there are
times when a reverse acting control output is needed. The D4-454 loop controller allows a loop
to operate as reverse acting. With a reverse acting loop, the output is driven in the opposite
direction of the error. For example, if SP > PV, then a reverse acting controller will decrease
the output to increase the PV.
Mx = -Ki * en + Mxn-1
M = -Kc * en + Kr(PVn-PVn-1) + Mxn
Square Root of the Process Variable
Square root is selected whenever the PV is from a device such as an orifice meter which requires
this calculation.
Error Squared Control
Whenever error squared control is selected, the error is calculated as:
en = (SP - PVn) * ABS(SP - PVn)
A loop using the error squared is less responsive than a loop using just the error, however, it
will respond faster with a large error. The smaller the error, the less responsive the loop. Error
squared control would typically be used in a PH control application.

Error Deadband Control

With error deadband control, no control action is taken if the PV is within the specified
deadband area around the setpoint. The error deadband is the same above and below the
setpoint.
Once the PV is outside of the error deadband around the setpoint, the entire error is used in
the loop calculation.
en = 0 “SP - Deadband_Below_SP < PV < SP - Deadband_Above_SP”
en = P - PVn “otherwise.”
The error will be squared first if both Error Squared and Error Deadband is selected.
Derivative Gain Limiting
When the coefficient of the derivative term, Kr, is a large value, noise introduced into the
PV can result in erratic loop output. This problem is corrected by specifying a derivative gain
limiting coefficient, Kd. Derivative gain limiting is a first order filter applied to the derivative
term computation, Yn, as shown below.
Ts
Y n = Yn-1 + * (PV n - Y n-1 )
Ts + ( Td )
Kd
Position Algorithm
M = Kc * en - Kr * (Yn-Yn-1) + Mx
Velocity Algorithm
DM = Kc * (en - en-1) + Ki * en - Kr * (Yn - 2 * Yn-1 + Yn-2)

Ten Steps to Successful Process Control

Controllers such as the D4-454 PLC provide sophisticated process control features. Automated
control systems can be difficult to debug, because a given symptom can have many possible
causes. We recommend a careful, step-by-step approach to bringing new control loops online:
Step 1: Know the Recipe
The most important is – how to produce your product. This knowledge is the foundation for
designing an effective control system. A good process recipe will do the following:
• Identify all relevant Process Variables, such as temperature, pressure, or flow rates, etc.,
which need precise control.
• Plot the desired Setpoint values for each process variables for the duration of one process
cycle.
Step 2: Plan Loop Control Strategy
This simply means choosing the method the machine will use to maintain control over the
Process Variables to follow their Setpoints. This involves many issues and trade-offs, such as
energy efficiency, equipment costs, ability to service the machine during production, and more.
You must also determine how to generate the Setpoint value during the process, and whether
a machine operator can change the SP.
Step 3: Size and Scale Loop Components
Assuming the control strategy is sound, it is still crucial to properly size the actuator and properly
scale the sensors.
• Choose an actuator (heater, pump. etc.) which matches the size of the load. An oversized
actuator will have an overwhelming effect on your process after a SP change. However, an
undersized actuator will allow the PV to lag or drift away from the SP after a SP change
or process disturbance.
• Choose a PV sensor which matches the range of interest (and control) for our process.
Decide the resolution of control you need for the PV (such as within 2!C), and make sure
the sensor input value provides the loop with at least 5 times that resolution (at LSB level).
However, an over-sensitive sensor can cause control oscillations, etc. The D4-454 provides
12–bit and 15–bit unipolar and bipolar data format options, and a 16–bit unipolar option.
This selection affects SP, PV, Control Output and Integrator sum.
Step 4: Select I/O Modules
After deciding the number of loops, PV variables to measure, and SP values, you can choose
the appropriate I/O module. In many cases, you will be able to share input or output modules,
or use an analog I/O combination module, among several control loops. The example shown
sends the PV and Control Output signals for two loops through the same set of modules.
DL06 CPU
Input V-memory Digital
Module Output
Loop 1 Data
Channel 1 PV SP OUT Channel 1 Process 1
Loop 2 Data
Channel 2 PV SP OUT Channel 2 Process 2
Channel 3
Channel 4

Step 5: Wiring and Installation

• After selection and procurement of all loop components and I/O module(s), you can
perform the wiring and installation. Refer to the wiring guidelines in Chapter 2 of this
manual. The most common wiring errors when installing PID loop controls are:
• Reversing the polarity of sensor or actuator wiring connections.
• Incorrect signal ground connections between loop components.
Step 6: Loop Parameters
After wiring and installation, choose the loop setup parameters. The easiest method for
programming the loop tables is using DirectSOFT (6.1 or later). This software provides PID
Setup using dialog boxes to simplify the task. Note: It is important to understand the meaning
of all loop parameters mentioned in this chapter before choosing values to enter.
Step 7: Check Open Loop Performance
With the sensor and actuator wiring done, and loop parameters entered, we must manually and
carefully check out the new control system using the Manual mode.
• Verify that the PV value from the sensor is correct.
• If it is safe to do so, gradually increase the control output up above 0%, and see if the PV
responds (and moves in the correct direction!).
Step 8: Loop Tuning
If the Open Loop Test (page 8–42) shows the PV reading is correct and the control output has
the proper effect on the process; you can follow the closed loop tuning procedure (see page
8–49). In this step, the loop is tuned so the PV automatically follows the SP.
Step 9: Run Process Cycle
If the closed loop test shows the PV will follow small changes in the SP, consider running an
actual process cycle. You will need to have completed the programming which will generate the
desired SP in real time. In this step, you may want to run a small test batch of product through
the machine, watching the SP change according to the recipe.
WARNING: Be sure the Emergency Stop and power-down provision is readily accessible,
in case the process goes out of control. Damage to equipment and/or serious injury to
personnel can result from loss of control of some processes.
Step 10: Save Parameters

When the loop tests and tuning sessions are complete, be sure to save all loop setup parameters
to disk.

PID Loop Setup

Some Things to Do and Know Before Starting
Have your analog module installed and operational before beginning the loop setup (refer to
the D4-ANLG-M Manual). The D4-454 PLC gets its PID loop processing instructions from
V-memory tables. There isn’t a PID instruction that can be used in RLL, such as a block,
to setup the PID loop control. Instead, the CPU reads the setup parameters from system
V-memory locations. These locations are shown in the table below for reference only; they can
be used in a RLL program if needed.
Address Setup Parameter Data Type Ranges Read/Write
V7640 Loop Parameter Octal V1200 – V7340 write
Table Pointer V10000-V17740
V7641 Number of Loops BCD 1 – 16 write
V7642 Loop Error Flags BITS 0 or 1 read
NOTE: It is recommended to set the V-memory used to store the PID Setup Parameters to retentive. If
power is removed from the CPU, these parameters will not be written to zero on power up.
PID Error Flags, V7642
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
PID ERROR Flags

The CPU reports any programming errors of the setup parameters in V7640 and V7641. It
does this by setting the appropriate bits in V7642 on program-to-run mode transitions.
If you use the DirectSOFT loop setup dialog box, its automatic range checking prohibits
possible setup errors. However, the setup parameters may be written using other methods such
as RLL, so the error flag register may be helpful in those cases. The following table lists the
errors reported in V7642.
As a quick check, if the CPU is in Run mode and V7642=0000, there are no programming
errors.
Bit Error Description (0 = no error, 1 = error)
0 The starting address (in V7640) is out of the lower V-memory range.
1 The starting address (in V7640) is out of the upper V-memory range.
2 The number of loops selected (in V7641) is greater than 16.
3 The loop table extends past (straddles) the boundary at V7377. Use an address closer to V1400.
4 The loop table extends past (straddles) the boundary at V36777. Use an address closer to V10000.

Establishing the Loop Table Size and Location

On a PROGRAM-to-RUN mode transition, the CPU reads the loop setup parameters as
pictured below. At that moment, the CPU learns the location of the loop table and the number
of loops it configures. Then during the ladder program scan, the PID Loop task uses the loop
data to perform calculations, generate alarms, and so on. There are some loop table parameters
the CPU will read or write on every loop calculation.
CPU Tasks V–Memory Space
User Data
READ/
Ladder WRITE LOOP
Program DATA
CONFIGURE/
MONITOR
PID Loop
Task
Setup Parameters
READ V7640, V7641
(at powerup)
DirectSOFT 6.1 or later Programming Software
NOTE: The D4-454 CPU’s PID algorithm requires DirectSOFT Version 6.1 (or later) and firmware version 1.0
(or later). See our website for more information: www.automationdirect.com.
V–Memory User Data

The Loop Table contains data for only the number of loops
that are selected. The address for the table is stored in V7641. V2000
LOOP #1
Each loop configuration occupies 32 words (0 to 37 octal) in V2037 32 words
the loop table. V2040 LOOP #2
32 words
For example, consider an application with 4 loops, and V2000 V2077
has been chosen as the starting location. The Loop Parameter LOOP #3
32 words
will occupy V2000 – V2037 for loop 1, V2040 – V2077 for LOOP #4
loop 2 and so on. Loop 4 occupies V2140 – V2177. 32 words
Determine the block of V-memory to be used for each PID loop. Besides being the beginning
of the PID parameter memory block, the first address will be the start of loop 1 parameters.
Remember, there are 32 words (0 to 37 octal) needed for each loop. Once you have determined
the beginning V-memory address to be used, you can setup and store the PID parameters either
directly in your RLL program or by the using PID Setup in DirectSOFT.
NOTE: Whether one or more loops are being setup, this block of V-memory will only be used for the PID loop
parameters, do not use this block of memory for anything else in your program.

Using DirectSOFT is the simplest way to setup the parameters. To setup the PID parameters,
the D4-454 must be powered up and connected to the programming computer. The
parameters can only be entered in PID setup when the PLC is in the Program mode. Once
the parameters have been entered and saved for each loop, changes made through the PID
setup can be made, but only in Program Mode. You can type the beginning address in the PID
Table Address dialog found when the PID Setup is opened in DirectSOFT. This can be seen
in the diagram below. After the address has been entered, the memory range will appear. Also,
entering the number of PID loops (1 to 16) will set the total V-memory range for the number
of loops entered. After the V-memory address has been entered, the necessary PID parameters
for a basic loop operation for each loop can be setup with the dialogs made available.
NOTE: Have an edited program open, then click on PLC > Setup > PID to access the Setup PID dialog.

Loop Table Word Definitions

These are the loop parameters associated with each of the loops available in the D4-454. The
address offset is in octal, to help you locate specific parameters in the loop table. For example,
if a table begins at V2000, then the location of the reset (integral) term is Addr+11, or V2011.
Do not use the Word # (in the first column) to calculate addresses.
Read
Word # Address+Offset Description Format on-the-
fly***
1 Addr + 0 PID Loop Mode Setting 1 Bits Yes
2 Addr + 1 PID Loop Mode Setting 2 Bits Yes
3 Addr + 2 Setpoint Value (SP) Word/Binary Yes
4 Addr + 3 Process Variable (PV) Word/Binary Yes
5 Addr + 4 Bias (Integrator) Value Word/Binary Yes
6 Addr + 5 Control Output Value Word/Binary Yes
7 Addr + 6 Loop Mode and Alarm Status Bits –
8 Addr + 7 Sample Rate Setting Word/Bcd Yes
9 Addr + 10 Gain (Proportional) Setting Word/Bcd Yes
10 Addr + 11 Reset (Integral) Time Setting Word/Bcd Yes
11 Addr + 12 Rate (Derivative) Time Setting Word/Bcd Yes
12 Addr + 13 PV Value, Low-low Alarm Word/Binary No*
13 Addr + 14 PV Value, Low Alarm Word/Binary No*
14 Addr + 15 PV Value, High Alarm Word/Binary No*
15 Addr + 16 PV Value, High-high Alarm Word/Binary No*
16 Addr + 17 PV Value, deviation alarm (YELLOW) Word/Binary No*
17 Addr + 20 PV Value, deviation alarm (RED) Word/Binary No*
18 Addr + 21 PV Value, rate-of-change alarm Word/Binary No*
19 Addr + 22 PV Value, alarm hysteresis setting Word/Binary No*
20 Addr + 23 PV Value, error deadband setting Word/Binary Yes
21 Addr + 24 PV low-pass filter constant Word/Bcd Yes
22 Addr + 25 Loop derivative gain limiting factor setting Word/Bcd No**
23 Addr + 26 SP value lower limit setting Word/Binary Yes
24 Addr + 27 SP value upper limit setting Word/Binary Yes
25 Addr + 30 Control output value lower limit setting Word/Binary No**
26 Addr + 31 Control output value upper limit setting Word/Binary No**
27 Addr + 32 Remote SP Value V-Memory Address Pointer Word/Hex Yes
28 Addr + 33 Ramp/Soak Setting Flag Bit Yes
29 Addr + 34 Ramp/Soak Programming Table Starting Address Word/Hex No**
30 Addr + 35 Ramp/Soak Programming Table Error Flags Bits No**
31 Addr + 36 PV auto transfer, channel number Word/Hex Yes
32 Addr + 37 Control output auto transfer, channel number Word/Hex Yes
* Read data only when alarm enable bit transitions from 0 to 1.
** Read data only on PLC Mode change.
*** Read on-the-fly means that the content of V-memory can be changed while the PID loop is in
operation.

PID Mode Setting 1 Bit Descriptions (Addr + 00)

The individual bit definitions of the PID Mode Setting 1 word (Addr+00) are listed in the
following table.
Bit PID Mode Setting 1 Description Read/Write Bit=0 Bit=1

0 Manual Mode Loop Operation request Write – 0 –> 1 request
1 Automatic Mode Loop Operation request Write – 0 –> 1 request
2 Cascade Mode Loop Operation request Write – 0 –> 1 request
3 Bumpless Transfer select Write Mode I Mode II
4 Direct or Reverse-Acting Loop select Write Direct Reverse
5 Position / Velocity Algorithm select Write Position Velocity
6 PV Linear / Square Root Extract select Write Linear Sq. root
7 Error Term Linear / Squared select Write Linear Squared
8 Error Deadband enable Write Disable Enable
9 Derivative Gain Limit select Write Off On
10 Bias (Integrator) Freeze select Write Off On
11 Ramp/Soak Operation select Write Off On
12 PV Alarm Monitor select Write Off On
13 PV Deviation alarm select Write Off On
14 PV rate-of-change alarm select Write Off On
Loop with CPU Loop Independent
15 Loop mode is independent from CPU mode when set Write
mode of CPU mode

PID Mode Setting 2 Bit Descriptions (Addr + 01)

The individual bit definitions of the PID Mode Setting 2 word (Addr+01) are listed in the
following table.
Bit PID Mode 2 Word Description Read/Write Bit=0 Bit=1

0 Input (PV) and Control Output Range Unipolar/Bipolar Write Unipolar Bipolar
select (See Notes 2 and 3)
1 Input/Output Data Format select (See Notes 2 and 3) Write 12 Bit 15 Bit
2 Analog Input filter Write Off O
3 SP Input limit enable Write Disable Enable
4 Integral Gain (Reset) units select Write Seconds Minutes
5 Select Auto tune PID algorithm Write Closed loop Open loop
6 Auto tune selection Write PID PI only (rate = 0)
Auto tune start Auto tune
7 Read/Write Force start
(See Note 1) Cancel/done
8 PID Scan Clock (internal use) Read – –
Input/Output Data Format 16-bit select
9 Write Not 16 bit Select 16 bit
(See Notes 2 and 3)
Select separate data format for input and output
10 Write Same format Separate formats
(See Notes 3 and 4)
Control Output Range Unipolar/Bipolar select
11 Write Unipolar Bipolar
(See Notes 3 and 43)
12 Output Data Format select (See Notes 3 and 4) Write 12 Bit 15 Bit
13 Output data format 16-bit select (See Notes 3 and 4) Write Not 16 bit Select16 bit
14–15 Reserved for future use – – –
NOTE 1: Bit 7 can be used to cancel Autotune mode by setting it to 0

NOTE 2: If the value in bit 9 is 0, then the values in bits 0 and 1 are read. If the value in bit 9 is 1, then the
values in bits 0 and 1 are not read, and bit 9 defines the data format (the range is automatically unipolar).
NOTE 3: If the value in bit 10 is 0, then the values in bits 0, 1 and 9 define the input and output ranges and
data formats (the values in bits 11, 12, and 13 are not read). If the value in bit 10 is 1, then the values in
bits 0, 1, and 9 define only the input range and data format, and bits 11, 12, and 13 are read and define the
output range and data format.
NOTE 4: If bit 10 has a value of 1 and bit 13 has a value of 0, then bits 11 and 12 are read and define the
output range and data format. If bit 10 and bit 13 each have a value of 1, then bits 11 and 12 are not read,
and bit 13 defines the data format, (the output range is automatically unipolar).

Mode/Alarm Monitoring Word (Addr + 06)

The individual bit definitions of the Mode / Alarm monitoring (Addr+06) word are listed in
the following table.
Bit Mode/Alarm Bit Description Read/Write Bit=0 Bit=1
0 Manual Mode Indication Read – Manual
1 Automatic Mode Indication Read – Auto
2 Cascade Mode Indication Read – Cascade
3 PV Input LOW–LOW Alarm Read Off On
4 PV Input LOW Alarm Read Off On
5 PV Input HIGH Alarm Read Off On
6 PV Input HIGH–HIGH Alarm Read Off On
7 PV Input YELLOW Deviation Alarm Read Off On
8 PV Input RED Deviation Alarm Read Off On
9 PV Input Rate-of-Change Alarm Read Off On
10 Alarm Value Programming Error Read – Error
11 Loop Calculation Overflow/Underflow Read – Error
12 Loop in Auto-Tune indication Read Off On
13 Auto-Tune error indication Read – Error
14–15 Reserved for Future Use – – –
Ramp/Soak Table Flags (Addr + 33)

The individual bit definitions of the Ramp/Soak Table Flag (Addr+33) word are listed in the
following table.
Bit Ramp/Soak Flag Bit Description Read/Write Bit=0 Bit=1

0 Start Ramp / Soak Profile Write – 0 –> Start
1 Hold Ramp / Soak Profile Write – 0 –> Hold
2 Resume Ramp / soak Profile Write – 0 –> Resume
3 Jog Ramp / Soak Profile Write – 0 –> Jog
4 Ramp / Soak Profile Complete Read – Complete
5 PV Input Ramp / Soak Deviation Read Off On
6 Ramp / Soak Profile in Hold Read Off On
7 Reserved Read – –
8–15 Current Step in R/S Profile Read Decode as byte (hex)
Bits 8–15 must be read as a byte to indicate the current segment number of the Ramp/Soak
generator in the profile. This byte will have the values 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E,
F, and 10, which represent segments 1 to 16 respectively. If the byte=0, then the Ramp/Soak
table is not active.

Ramp/Soak Table Location (Addr + 34)

Each loop that you configure has the option of using a built-in Ramp/Soak generator
dedicated to that loop. This feature generates SP values that follow a profile. To use the Ramp
Soak feature, you must program a separate table of 32 words with appropriate values. The
DirectSOFT dialog box makes this easy to do.
In the loop table, the Ramp/Soak Table Pointer at Addr+34 must point to the start of the
ramp/soak data for that loop. This may be anywhere in user memory, and does not have to
adjoin to the Loop Parameter table, as shown to the left. Each R/S table requires 32 words,
regardless of the number of segments programmed.
The ramp/soak table parameters are defined in the table below. Further details are in the section
on Ramp/Soak Operation in this chapter.
Addr Addr
Step Description Step Description
Offset Offset
+ 00 1 Ramp End SP Value + 20 9 Ramp End SP Value
+ 01 1 Ramp Slope + 21 9 Ramp Slope
V–Memory Space + 02 2 Soak Duration + 22 10 Soak Duration
User Data
+ 03 2 Soak PV Deviation + 23 10 Soak PV Deviation
V2000 LOOP #1
V2037 32 words + 05 3 Ramp Slope + 25 11 Ramp Slope
LOOP #2 + 06 4 Soak Duration + 26 12 Soak Duration
32 words
V3000 Ramp/Soak #1
32 words
+ 12 6 Soak Duration + 32 14 Soak Duration
V2034 = 3000 Octal
Pointer to R/S table
Ramp/Soak Table Programming Error Flags (Addr + 35)

The individual bit definitions of the Ramp/Soak Table Programming Error Flags word
(Addr+35) are listed in the following table. Further details are given in the PID Loop Mode
section and in the PV Alarm section later in chapter 8.
Bit R/S Error Flag Bit Description Read/Write Bit=0 Bit=1
0 Starting Addr out of lower V-memory range Read – Error
1 Starting Addr out of upper V-memory range Read – Error
4 Starting Addr in System Parameter V-memory Range Read – Error

Configure the PID Loop

Once the PID table is established in V-memory, configuring the PID loop continues with
the DirectSOFT PID setup configuration dialog. You will need to check and fill in the data
required to control the PID loop. Select Configure and the following dialog will appear for this
process.
Select the Algorithm Type

Chose either Position or Velocity. The default algorithm is Position. This is the choice for most
applications which include heating and cooling loops as well as most position and level control
loops. A typical velocity control will consist of a process variable such as a flow totalizer in a
flow control loop.
Enter the Sample Rate
The main tasks of the CPU fall into categories as shown to the
Read
right. The list represents the tasks done when the CPU is in Inputs
Run Mode, on each PLC scan. Note that PID loop calculations
occur after the ladder logic task. Service
Peripherals
The sample rate of a control loop is simply the frequency of Ladder

the PID calculation. Each calculation generates a new control PLC
Program
output value. With the DL405 CPU, you can set the sample Scan
rate of a loop from 50 ms to 99.99 seconds. Most loops do not Calculate
PID Loops
require a fresh PID calculation on every PLC scan. Some loops
may need to be calculated only once in 1000 scans.
Internal
Enter 0.05 sec., or the sample rate of your choice, for each Diagnostics
loop, and the CPU automatically schedules and executes PID

calculations on the appropriate scans. Write
Outputs
NOTE: If more than 4 loops are programmed, enter a minimum of 0.1 second.

Select Forward/Reverse
It is important to know which direction the control output will respond to the error (SP-
PV), either forward or reverse. A forward (direct) acting control loop means that whenever the
control output increases, the process variable will also increase. The control outputs of most
PID loops are forward acting, such as a heating control loop. An increase in heat applied will
increase the PV (temperature).
A reverse acting control loop is one where an increase in the control output results in a decrease
in the PV. A common example of this would be a refrigeration system, where an increase in the
cooling input causes a decrease in the PV (temperature).
The Transfer Mode
Choose either Bumpless I or Bumpless II to provide a smooth transition of the control output
from Manual Mode to Auto Mode. Choosing Bumpless I will set the SP equal to the PV when
the control output is switched from Manual to Auto. If this is not desired, choose Bumpless II.
The characteristics of Bumpless I and II transfer types are listed in the chart below. Note that
their operation also depends on which PID algorithm you are using, the position or velocity
form of the PID equation. Note that you must use Bumpless Transfer type I when using the
velocity form of the PID algorithm.
Transfer PID Algorithm Manual-to-Auto Auto-to-Cascade
Transfer Type Select Bit 3 Transfer Action Transfer Action
Position Forces Bias = Control Output Forces Major Loop Output =
Bumpless Forces SP = PV Minor Loop PV
Transfer I 0
Velocity Forces SP = PV Forces Major Loop Output =
Minor Loop PV
Bumpless Position Forces Bias = Control Output None
1
Transfer II Velocity None None
The transfer type can also be selected in an RLL program by setting bit 3 of PID Mode 1, V+00
setting as shown.
PID Mode 1 Setting V+00
Bumpless Transfer I/II Select
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
SP/PV & Output Format

This block allows you to select either Common format or Independent format. Common format
is the default and is most commonly used. With this format, both SP/PV and Output will have
the same data structure. Both will have the same number of bits and either bipolar or unipolar.
If Independent format is selected, the data structure selections will be grayed out. The reason
for this is that they become independently selectable in the SP/PV and the Output dialogs.
Common Data Format
Select either Unipolar data format (which is positive data only) in 12 bit (0 to 4095), 15 bit (0
to 32767), or 16 bit (0 to 65535) format, or Bipolar data format, which ranges from negative
to positive (-4095 to 4095 or -32767 to 32767) and requires a sign bit. Bipolar selection
displays input/output as magnitude plus sign, not two’s complement. The bipolar selection is
not available when 16-bit data format is selected.

Setpoint V+02 Loop Control Output V+05

+ Calculation
–
PID Mode 2 Setting V+01 Process Variable V+03

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Data formats LSB

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
00 12 bit unipolar 0 to 0FFF (0 to 4095)
Select data
01 12 bit bipolar 0 to 0FFF, 8FFF to 8001
format using
(0 to 4095, *–4095 to 4095)
bits 0 and 1.
10 15 bit unipolar 0 to 32767
11 15 bit bipolar 0 to 7FFF, FFFF to 8001

(0 to 32767, *–32767 to 32767)
= sign bit
* Magnitude plus sign
The data format determines the numerical interface between the PID loop and the PV sensor
and the control output device. This selects the data format for both the SP and the PV.
Loop Mode
Loop Mode is a special feature that allows the PID loop controller to perform closed-loop
control while the CPU is in the Program Mode. Careful thought must be taken before using
this feature called Independent of CPU mode in the dialog. Before continuing with the PID
setup, a knowledge of the three PID loop modes will be helpful.
The DL405 provides the three standard control modes: Manual, Automatic, and Cascade. The
sources of the three basic variables SP, PV and control output are different for each mode.
In Manual Mode, the loop is not executing PID calculations (however, loop alarms are still
active). With regard to the loop table, the CPU stops writing values to location V+05 (control
output) for that loop. It is expected that an operator or other intelligent source is manually
controlling the output by observing the PV and writing data to the control output as necessary
to keep the process under control. The drawing below shows the equivalent schematic diagram
of manual mode operation.
Input from Operator Manual
Control Output V+05
Loop
Calculation
Auto
In Automatic Mode, the loop operates normally and generates new control output values. It
calculates the PID equation and writes the result in location V+05 every sample period of that
loop. The equivalent schematic diagram is shown below.
Input from Operator Manual
Control Output V+05
Loop
Calculation
Auto

In Cascade Mode, the loop operates as it does in Automatic Mode, with one important
difference. The data source for the SP changes from its normal location at V+02 to using
the control output value, V+05, from another loop. So in Auto or Manual modes, the loop
calculation uses the data at V+02. In Cascade Mode, the loop calculation reads the control
output from another loop’s parameter table, V+05.
Another loop Cascaded loop
Control Output V+05 Cascade

Loop
Calculation Control Output
Setpoint Loop
+ Calculation
Normal SP V+02 –
Auto/Manual Process Variable
As pictured below, a loop can be changed from one mode to another, but cannot go from
Manual Mode directly to Cascade, or vice versa. This mode change is prohibited because a loop
would be changing two data sources at the same time, and could cause a loss of control.
Manual Automatic Cascade
Once the CPU is operating in the Run Mode, the normal operation of the PID loop controller
is to read the loop data and perform calculations on each scan of the RLL program. When the
CPU is placed in the Program Mode, the RLL program halts operation and all PID loops are
automatically put into the Manual Mode. The PID parameters can then be changed if desired.
Similarly, by placing the CPU in the Run mode, the PID loops are returned to the operational
mode which they were previously in, i.e., Manual, Automatic and Cascade. With this selection
you automatically affect the modes by changing the CPU mode.
CPU Modes:
Mode change
Program Run
Loop Mode Linking
0 = loop follows PLC mode PID Mode 1 Setting V+00

1 = loop is independent
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
of PLC mode
Loop
Modes:
Mode change Mode change
Manual Automatic Cascade
If bit 15 is set to one, then the loops will run independently of the CPU mode. It is like having
two independent processors in the CPU... one is running the RLL program and the other is
running the process loops.

Having the ability to run loops independently of the RLL program makes it feasible to make a
ladder logic change while the process is still running. This is especially beneficial for large-mass
continuous processes that are difficult or costly to interrupt. The Independent of CPU is the
feature used for this.
If you need to operate the PID loops while the RLL program is halted, in Program Mode,
either select the Independent of CPU mode in the dialog or edit your program to set and reset
bit 15 of PID Mode 1 word (V+00) in your RLL program. If the bit is set to a zero, the loop
will follow the CPU mode; then, when the CPU is placed in the Program Mode, all loops will
be forced into the Manual Mode.
When Independent of CPU mode is used, you should also set the PV to be read directly from
an analog input module. This can easily be done in the PID setup dialog, SP/PV.
The SP/PV dialog has a block entitled Process

Variable. There is a block within this block called
Auto Transfer From (from analog input) with the
information grayed out. Checking the box to the
left of the Auto Transfer From will highlight the
information. Select I/O Module then enter the slot
number in which the input module resides. Next,
select the analog input channel of your choice.
The second choice is V-Memory. When this is
selected, the V-memory address from where the
PV is transferred must be specified.
Whichever method of auto transfer is used, it is
recommended to check the Enable Filter Factor (a
low pass filter) and specify the coefficient.
You should also select the analog output for the control output
to be transferred to. This is done in the PID setup Output dialog
shown here. The block of information in this dialog is grayed-
out until the box next to Auto transfer to I/O module is checked.
Once checked, enter the slot number where the output module
is residing and then enter the analog output channel number.
NOTE: To make changes to any loop table parameters, the PID loop must be in Manual mode and the
PLC must be stopped. If you have selected to operate the PID loop independent of the CPU mode,
then you must take certain steps to make it possible to make loop parameter changes. You can
temporarily make the loops follow the CPU mode by changing bit 15 to 0. Then, you will be able to
place the loop into Manual Mode using DirectSOFT. After you change the loop’s parameter settings,
restore bit 15 to 1.

You may optionally configure each loop to access its analog I/O (PV and control output) by
placing proper values in the associated loop table registers in your RLL program. The following
figure shows the loop table parameters at V+36 and V+37 and their auto transfer role to access
the analog values directly.
Setpoint V +02 Error Loop Control output V+05

Calculation
Process variable V+03
Loop Table
V2036 0X XX Base/Slot /Channel number for PV
V2037 0X XX Base/Slot/Channel number for Output
XX 0X
Channel number 1 to 4
Slot number 1 to 4
Base 0
When these loop table parameters are programmed directly, a value of 0102 in register V2036
directs the loop controller to read the PV data from channel 1 of the analog input. A value of
0000 in either register tells the loop controller not to access the corresponding analog value
directly. In that case, ladder logic must be used to transfer the value between the analog input
and the loop table.
NOTE: When auto transfer to/from I/O is used, the analog data for all of the channels on the analog module
cannot be accessed by any other method, i.e., pointer or multiplex.

SP/PV Addresses
An SP/PV dialog will be made available to setup how the setpoint (SP) and the process variable
(PV) will be used in the loop. If this loop is the minor loop of a cascaded pair, enter that control
output address in the Remote SP from Cascaded Loop Output area. It is sometimes desirable to
limit the range of setpoint values allowed to be entered. To activate this feature, check the
box next to Enable Limiting. This will activate the Upper and Lower fields for the values to be
entered. Set the limits around the SP value to prevent an operator from entering a setpoint
value outside of a safe range. The Square root box is only checked for certain PID loops, such as
a flow control loop. If the Auto transfer from I/O module is selected, a first-order low-pass filter
can be used by checking the Enable Filter box. The filter coefficient is user specified. The use of
this filter is recommended during closed loop auto-tuning. If the Independent format had been
checked previously, make the Data format selections here.
NOTE: The SP/PV dialog can be left as it first appears for basic PID operation.

Set Control Output Limits

Another dialog that will be available in the PID setup will be the Output dialog. The control
output address, V+05, (determined by the PID loop table beginning address) will be in view.
Enter the output range limits, Upper Limit and Lower Limit, that will meet the requirement of
the process and which will agree with the data format that has been selected. For a basic PID
operation using a 12-bit output module, set the Upper Limit to 4095 and leave the Lower Limit
set to 0.
Check the box next for Auto transfer to I/O module if there is a need to send the control output
to a certain analog output module, as in the case of using the Loop Mode Independent of
CPU Mode; otherwise, the PID output signal cannot control the analog output when the PLC
is not in RUN Mode. If the Auto transfer to I/O module feature is checked, all channels of the
module must be used for PID control outputs. If Independent format has been previously
chosen, the Output Data Format will need to be setup here, that is, select Unipolar or Bipolar
format and the bit structure. This area is not available and is grayed out if Common format has
been chosen (see page 8-26).
WARNING: If the Upper Limit is set to zero, the output will never get above zero. In effect,
there will be no control output. The default value is zero, so this value MUST be changed.

Enter PID Parameters

Another PID setup dialog, Tuning, is for entering the PID parameters shown as: Gain
(Proportional Gain), Reset (Integral Gain) and Rate (Derivative Gain).
Recall the position and velocity forms of the PID loop equations which were introduced earlier.
The equations basically show the three components of the PID calculation: Proportional Gain
(P), Integral Gain (I) and Derivative Gain (D). The following diagram shows a form of the
PID calculation in which the control output is the sum of the proportional gain, integral
gain and derivative gain. With each calculation of the loop, each term receives the same
error signal value.
Loop Calculation
P
Setpoint Error T erm + Control Output
+ I +
– D +
Process Variable
The P, I and D gains are 4-digit BCD numbers with values from 0000 to 9999. They contain
an implied decimal point in the middle, so the values are actually 00.00 to 99.99. Some gain
values have units – Proportional gain has no unit, Integral gain may be selected in seconds or
in minutes, and Derivative gain is in seconds.
Gain (Proportional Gain) – This is the most basic gain of the three. Values range from 0000
to 9999, but they are used internally as xx.xx. An entry of “0000” effectively removes the
proportional term from the PID equation. This accommodates applications which need
integral-only loops.
Reset (Integral Gain) – Values range from 0001 to 9998, but they are used internally as xx.xx.
An entry of “0000” or “9999” causes the integral gain to be “infinity”, effectively removing
the integrator term from the PID equation. This accommodates applications which need
proportional-only loops. The units of integral gain may be either seconds or minutes, as shown
in the above dialog.
Rate (Derivative Gain) – Values can be entered in the range of 0001 to 9999, but they are
used internally as XX.XX. An entry of “0000” allows removal of the derivative term from
the PID equation (a common practice). This accommodates applications which require only
proportional and/or integral loops. Most control loops will operate as a PI loop.

NOTE: You may elect to leave the tuning dialog blank and enter the tuning parameters in the DirectSOFT
PID View.
Derivative Gain Limiting

The derivative gain (rate) has an optional gain-limiting feature. This is provided because the
derivative gain reacts badly to PV signal noise or other causes of sudden PV fluctuations. The
function of the gain-limiting is shown in the diagram below.
Loop Calculation
Proportional
P
Control
Setpoint Error T erm Integral + Output
+ I +
– +
Derivative
D 0
Process Variable
Derivative, 1
gain-limited
Loop Table P ID Mode 1 Setting V+00

V+25 00XX Derivative Gain Limit
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Derivative gain limit select

The gain limit can be particularly useful during loop tuning. Most loops can tolerate only a
little derivative gain without going into uncontrolled oscillations.
If this option is checked, a Limit from 0 to 20 must also be entered.
NOTE: When first configuring a loop, it’s best to use the standard error term until after the loop is tuned.
Once the loop is tuned, you will be able to tell if these functions will enhance control. The Error Squared and/
or Enable Deadband can be selected later in the PID setup. Also, values are not required to be entered in the
Tuning dialog, but they can be set later in the DirectSOFT PID View.
Error Term Selection

The error term is internal to the CPUs PID loop controller, and is generated again in each
PID calculation. Although its data is not directly accessible, you can easily calculate it by
subtracting: Error = (SP–PV). If the PV square-root is enabled, then: Error = SP–√PV.
In any case, the size of the error and algebraic sign determine the next change of the
control output for each PID calculation.
Error Squared – When selected, the squared error function simply squares the error term (but
preserves the original algebraic sign), which is used in the calculation. This affects the Control
Output by diminishing its response to smaller error values, but maintaining its response to
larger errors. Some situations in which the error squared term might be useful:
• Noisy PV signal – using a squared error term can reduce the effect of low-frequency
electrical noise on the PV, which will make the control system jittery. A squared error
maintains the response to larger errors.
• Non-linear process – some processes (such as chemical pH control) require non-linear
controllers for best results. Another application is surge tank control, where the Control
Output signal must be smooth.

Enable Deadband – When selected, the enable deadband function takes a range of small error
values near zero, and simply substitutes zero as the value of the error. If the error is larger than
the deadband range, then the error value is used normally.
Freeze Bias
The term reset windup refers to an undesirable characteristic of integrator behavior which
occurs naturally under certain conditions. Refer to the figure below. Suppose the PV signal
becomes disconnected, and the PV value goes to zero. While this is a serious loop fault, it is
made worse by reset windup. Notice the bias (reset) term keeps integrating normally during
the PV disconnect, until its upper limit is reached. When the PV signal returns, the bias value
is saturated (windup) and takes a long time to return to normal. The loop output consequently
has an extended recovery time. Until recovery, the output level is wrong and causes further
problems.
PV
PV loss PV loss
0
Reset windup Freeze bias enabled
Bias
Output
Recovery time Recovery time
In the second PV signal loss episode in the figure, the freeze bias feature is enabled. It causes the
bias value to freeze when the control output goes out of bounds. Much of the reset windup is
thus avoided, and the output recovery time is much less.
For most applications, the freeze bias feature will work with the loop as described above. It is
suggested to enable this feature by selecting it in the dialog. Bit 10 of PID Mode 1 Setting
(V+00) word can also be set in RLL.
NOTE: The freeze bias feature stops the bias term from changing when the control output reaches the end of
the data range. If you have set limits on the control output other than the range (i.e, 0–4095 for a unipolar/12-
bit loop), the bias term still uses the end of range for the stopping point and bias freeze will not work.

Setup the PID Alarms

Although the setup of the PID alarms is optional, you surely would not want to operate a
process without monitoring it. The performance of a process control loop may generally be
measured by how closely the process variable matches the setpoint. Most process control loops
in industry operate continuously, and will eventually lose control of the PV due to an error
condition. Process alarms are vital in early discovery of a loop error condition and can alert
plant personnel to manually control a loop or take other measures until the error condition has
been repaired.
The alarm thresholds are fully programmable, and each type of alarm may be independently
enabled and monitored. The following diagram shows the Alarm dialog in the PID setup which
simplifies the alarm setup.
Monitor Limit Alarms

Checking this box will allow all of the PV limit alarms to be monitored once the limits are
entered. The PV absolute value alarms are organized as two upper and two lower alarms. The
alarm status is false as long as the PV value remains in the region between the upper and lower
alarms, as shown below. The alarms nearest the safe zone are named High Alarm and Low
Alarm. If the loop loses control, the PV will cross one of these thresholds first. Therefore, you
can program the appropriate alarm threshold values in the loop table locations shown below to
the right. The data format is the same as the PV and SP (12-bit or 15-bit). The threshold values
for these alarms should be set to give an operator an early warning if the process loses control.
High–high Alarm Loop Table
High Alarm V+16 XXXX High-high Alarm
PV V+15 XXXX High Alarm

V+14 XXXX Low Alarm
Low Alarm
V+13 XXXX Low-low Alarm
Low–low Alarm
NOTE: The Alarm dialog can be left as it first appears, without alarm entries. The alarms can then be setup
in the DirectSOFT PID View.

If the process remains out of control for some time, the PV will eventually cross one of the
outer alarm thresholds, named High-high alarm and Low-low alarm. Their threshold values
are programmed using the loop table registers listed above. A High-high or Low-low alarm
indicates a serious condition exists, and needs the immediate attention of the operator.
The PV Absolute Value Alarms are reported in bits 3, 4, 5 PID Mode and Alarm Status V+06
and 6 of the PID Mode and Alarm Status word (V+06) in
the loop table, as shown to the right. We highly recommend Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
using ladder logic to monitor these bits. The bit-of-word High-high Alarm
instructions make this easy to do. Additionally, you can High Alarm
Low Alarm
monitor PID alarms using DirectSOFT. Low-low Alarm
PV Deviation Alarms
The PV Deviation Alarms monitor the PV deviation with respect to the SP value. The deviation
alarm has two programmable thresholds, and each threshold is applied equally above and below
the current SP value. In the figure below, the smaller deviation alarm is called the Yellow
Deviation, indicating a cautionary condition for the loop. The larger deviation alarm is called
the Red Deviation, indicating a strong error condition for the loop. The threshold values use
the loop parameter table locations V+17 and V+20 as shown.
Red Deviation Alarm Red

Yellow Deviation Alarm Yellow Loop Table
SP Green V+17 XXXX Yellow Deviation Alarm
V+20 XXXX Red Deviation Alarm
Yellow Deviation Alarm Yellow
Red Deviation Alarm
Red
The thresholds define zones, which fluctuate with the SP value. The green zone which surrounds
the SP value represents a safe (no alarm) condition. The yellow zones lie outside the green zone,
and the red zones are beyond those.
The PV Deviation Alarms are reported in bits 7 and 8 of the PID Mode and Alarm Status V+06
PID Mode and Alarm Status word (V+06) in the loop table,
as shown to the right. We highly recommend using ladder Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
logic to monitor these bits. The bit-of-word instructions Red Deviation

make this easy to do. Additionally, you can monitor PID Yellow Deviation
alarms using DirectSOFT.
The PV Deviation Alarm can be independently enabled and disabled from the other PV alarms,
using bit 13 of the PID Mode 1 Setting V+00 word.
Remember the alarm hysteresis feature works in conjunction with both the deviation and
absolute value alarms, and is discussed at the end of this section.
NOTE: PID deviation alarm only works in Auto mode.

PV Rate-of-Change Alarm
An excellent way to get an early warning of a process fault is to monitor the rate-of-change of
the PV. Most batch processes have large masses and slowly-changing PV values. A relatively
fast-changing PV will result from a broken signal wire for either the PV or control output, a
SP value error, or other causes. If the operator responds to a PV Rate-of-Change Alarm quickly
and effectively, the PV absolute value will not reach the point where the material in process
would be ruined.
The D4-454 loop controller provides a programmable PV Rate-of-Change Alarm, as shown
below. The rate-of-change is specified in PV units change per loop sample time. This value is
programmed into the loop table location V+21.
Loop Table
PV slope OK PV slope excessive
V+21 XXXX PV Rate-of-Change Alarm
PV
PID Mode and Alarm Status V+06
rate-of-change alarm
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Sample time Sample time
PV Rate of
Change Alarm
As an example, suppose the PV is the temperature for your process, and you want an alarm
whenever the temperature changes faster than 15 degrees/minute. The PV counts per degree
and the loop sample rate must be known. Then, suppose the PV value (in V+03 location)
represents 10 counts per degree, and the loop sample rate is 2 seconds. Use the formula below
to convert our engineering units to counts/sample period:
15 degrees 10 counts / degree 150
Alarm Rate-of-Change = X = = 5 counts / sample period
1 minute 30 loop samples / min. 30
From the calculation result, you would program the value 5 in the loop table for the rate-of-
change. The PV Rate-of-Change Alarm can be independently enabled and disabled from the
other PV alarms, using bit 14 of the PID Mode 1 Setting V+00 word.
The alarm hysteresis feature (discussed next) does not affect the Rate-of-Change Alarm.

PV Alarm Hysteresis
The PV Absolute Value Alarm and PV Deviation Alarm are programmed using threshold
values. When the absolute value or deviation exceeds the threshold, the alarm status becomes
true. Real-world PV signals have some noise on them, which can cause some fluctuation in
the PV value in the CPU. As the PV value crosses an alarm threshold, its fluctuations cause
the alarm to be intermittent and annoy process operators. The solution is to use the PV Alarm
Hysteresis feature.
The PV Alarm Hysteresis amount is programmable from 1 to 200 (binary/decimal). When
using the PV Deviation Alarm, the programmed hysteresis amount must be less than the
programmed deviation amount. The figure below shows how the hysteresis is applied when the
PV value goes past a threshold and descends back through it.
Alarm threshold
Hysteresis
Loop Table
PV
V+22 XXXX PV Alarm Hysteresis
Alarm 1
0
The hysteresis amount is applied after the threshold is crossed, and toward the safe zone. In this
way, the alarm activates immediately above the programmed threshold value. It delays turning
off until the PV value has returned through the threshold by the hysteresis amount.
Alarm Programming Error
The PV Alarm threshold values must have certain
mathematical relationships to be valid. The
PID Mode and Alarm Status V+06
requirements are listed below. If not met, the Alarm
Programming Error bit will be set, as indicated to Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
the right.
Alarm Programming Error
• PV Absolute Alarm value requirements:
Low-low < Low < High < High-high
• PV Deviation Alarm requirements:
Yellow < Red
Loop Calculation Overflow/Underflow Error PID Mode and Alarm Status V+06
This error occurs whenever the output reaches its Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
upper or lower limit and the PV does not reach the

setpoint. A typical example might be when a valve Loop Calculation Overflow/Underflow Error
is stuck, the output is at its limit, but the PV has
not reached setpoint.
NOTE: Overflow/underflow can be alarmed in PID View. The optional C-more operator interface panel
(see the automationdirect.com website) can also be setup to read these error bits using the PID Faceplate
templates.

Ramp/Soak
R/S (Ramp/Soak) is the last dialog available in the PID setup. The basic PID does not require
any entries to be made in order to operate the PID loop. Ramp/Soak will be discussed in
another section.
Complete the PID Setup

Once you have filled in the necessary information for the basic PID setup, the configuration
should be saved. The icons on the Setup PID dialog will allow you to save the configuration to
the PLC and to disk. The save to icons have the arrow pointing to the PLC and disk. The read
from icons have the arrows pointing away from the PLC and disk.
An optional feature is available with the Doc tab in the Setup PID window. You enter a
name and description for the loop. This is useful if there is more than one PID loop in your
application.
Save to disk
Save to PLC
NOTE: It is good practice to save your project after setting up the PID loop by selecting File from the menu
toolbar, then Save project > to disk. In addition to saving your entire project, all the PID parameters are also
saved.

PID Loop Tuning

Once you have set up a PID loop, it must be tuned in order for it to work. The goal of loop
tuning is to adjust the loop gains so the loop has optimal performance in dynamic conditions.
The quality of a loop’s performance may generally be judged by how well the PV follows the
SP after a SP step change. It is important to keep in mind that understanding the process is
fundamental to getting a well designed control loop. Sensors must be in appropriate locations
and valves must be sized correctly with appropriate trim. PID control does not have typical
values. There isn’t one control process that is identical to another.
Manual Tuning vs. Auto Tuning
You may enter the PID gain values to tune your loops (manual tuning), or you can rely on the
PID processing engine in the CPU to automatically calculate the gain values (auto tuning).
Most experienced process engineers will have a favorite method; the D4-454 will accommodate
either preference. The use of auto tuning can eliminate much of the trial-and-error of the
manual tuning approach, especially if you do not have a lot of loop tuning experience. However,
performing the auto tuning procedure will get the gains close to optimal values, but additional
manual tuning can get the gain values to their optimal values.
WARNING: Only authorized personnel fully familiar with all aspects of the process should
make changes that affect the loop tuning constants. Using the loop auto tune procedures
will affect the process, including inducing large changes in the control output value. Make
sure you thoroughly consider the impact of any changes to minimize the risk of injury to
personnel or damage to equipment. The auto tune in the D4-454 is not intended to be used
as a replacement for your process knowledge.
Open-Loop Test
Whether you use manual or auto tuning, it is very important to verify basic characteristics of a
newly-installed process before attempting to tune it. With the loop in Manual Mode, verify the
following items for each new loop.
• Setpoint – verify that the SP source can generate a setpoint. Put the PLC in Run Mode
and leave the loop in Manual Mode, then monitor the loop table location V+02 to see the
SP value(s). (If you are using the ramp/soak generator, test it now).
• Process Variable – verify that the PV value is an accurate measurement, and the PV data
arriving in the loop table location V+03 is correct. If the PV signal is very noisy, consider
filtering the input either through hardware (RC low-pass filter), or using the filter in this
chapter.
• Control Output – if it is safe to do so, manually change the output a small amount
(perhaps 10%) and observe its affect on the process variable. Verify the process is direct-
acting or reverse acting, and check the setting for the control output (inverted or non-
inverted). Make sure the control output upper and lower limits are not equal to each other.
• Sample Rate – while operating open-loop, this is a good time to find the ideal sample rate
(see Configure the PID Loop on page 8-25). However, if you are going to use auto tuning,
the auto tuning procedure will automatically calculate the sample rate in addition to the
PID gains.

Manual Tuning Procedure

It is not necessary to try to obtain the best values for the P, I and D parameters in the PID loop
by trial and error. Following is a typical procedure for tuning a temperature control loop which
you may use to tune your loop.
Monitor the values of SP, PV and CV with a loop trending instrument or use the PID View
feature in DirectSOFT (see page 8-49).
NOTE: We recommend using the PID View Tuning and Trending window to select manual for the vertical
scale feature, for both SP/PV area and Bias/Control Output areas. The auto scaling feature would otherwise
change the vertical scale on the process parameters and add confusion to the loop tuning process.
• Adjust the gains so the Proportional Gain = 0.5 or 1.0 (1.0 is a good value based on
experience), Integral Gain = 9999 (this basically eliminates reset) and Derivative Gain =
0000. This disables the integrator and derivative terms, and provides some proportional
gain.
• Check the bias value in the PID View and set it to zero.
• Set the SP to a value equal to 50% of the full range.
• Now, select Auto Mode. If the loop will not stay in Auto Mode, check the troubleshooting
tips at the end of this chapter. Allow the PV to stabilize around the 50% point of the
range.
60% here error
10% of Over-damped PV response

SP range
SP 50% here
PV
• Change the SP to the 60% point of the range.

The response may take awhile, but you will see that there isn’t any oscillation. This response is
not desirable since it takes a long time to correct the error; also, there is a difference between
the SP and the PV.
60% here
10% of
SP range
PV response
SP 50% here
PV
• Increase the Proportional gain, for example to 2.0. The control output will be greater and
the response time will be quicker. The trend should resemble the figure below.
• Increase the Proportional gain in small increments, such as 4, 6, 7, etc., until the control
output response begins to oscillate. This is the Proportional gain that should be recorded.

Error
60% here
50% here
SP
PV
• Now, return the Proportional gain to the stable response; for example, 9.7. The error, SP-PV, should
be small, but not at zero.
• Next, add a small amount of Integral gain (reset) in order for the error to reach zero. Begin by using
80 seconds (adjust in minutes if necessary). The error should get smaller.
• Set the Integral gain to a lower value, such as 50 for a different response. If there is no response,
continue to decrease the reset value until the response becomes unstable. See the figure below.
60% here
10% of
SP range
SP 50% here
PV Under-damped PV response
• For discussion, let us say that a reset value of 35 made the control output unstable. Return the
reset value to the stable value, such as 38. Be careful with this adjustment since the oscillation can
destroy the process.
• The control output response should be optimal now, without a Derivative gain. The example
recorded values are: Proportional gain = 9.7 and Integral gain = 38 seconds. Note that the error has
been minimized.
Minimum Oscillations
Shortest response time
The foregone method is the most common method used to tune a PID loop. Derivative gain
is almost never used in a temperature control loop. This method can also be used for other
control loops, but other parameters may need to be added for a stable control output.

Test your loop for a high PV of 80% and again for a low PV of 20%, and correct the values if
necessary. Small adjustments of the parameters can make the control output more precise or
more unstable. It is sometimes acceptable to have a small overshoot to make the control output
react quicker.
The derivative gain can be helpful for those control loops which are not controlling temperature.
For these loops, try adding a value of 0.5 for the derivative gain and see if this improves the
control output. If there is little or no response, increase the derivative by increments of 0.5 until
there is an improvement to the output trend. Recall that the derivative gain reacts with a rate
of change of the error.

Alternative Manual Tuning Procedures by Others

The following tuning procedures have been extracted from various publications about PID
process control. These procedures are for comparison to the procedure in this manual.
Tuning PID Controllers

Two-Mode Simple Method - – for P-I controllers
1. Turn off reset and set the gain to a small value (0.5 - 1.0).
2. Increase gain until cycling starts, then decrease gain slightly.
3. Make setpoint changes to observe offset (error).
4. Increase reset to eliminate offset (error).
5. Repeat steps 2 through 4 until you obtain the largest gain and reset consistent
with the criteria of the control desired, i.e., offset, overshoot, stability.
Zeigler-Nichols Method– “Quarter amplitude decay”
1. Turn off reset and rate; set the proportional gain to a fairly large value.
2. Make a small setpoint change and observe how the controlled variable cycles.
3. Adjust the gain until the cycle is self-sustaining, and of constant
amplitude; this value is the ultimate gain (Gu).
4. Measure the period of cycling in minutes. This is the ultimate period (Pu).
5. Calculate the controller adjustments as follows:
P only: G = Gu/2
P & I : = Gu/2.2
G
Ti = 1.2/Pu (repeats/minute)
P-I-D: G = Gu/1.6
Ti = 2.0/Pu (repeats/minute)
Td = Pu/8.0 (minutes)
Pessen Method
1. Follow the procedure described above (Zeigler-Nichols) to
determine the ultimate gain and ultimate period.
2. Apply the formulas below.
For no overshoot during startup:
G = Gu/5.0
Ti = 3/Pu (repeats/minute)
Td = Pu/2 (minutes)
For some overshoot, but better response to disturbances:
G = Gu/3
Ti = 3/Pu (repeats/minute)
Td = Pu/3 (minutes)

Auto Tuning Procedure

The auto tuning feature for the D4-454 loop controller will only run once each time it is
enabled in the PID table. Therefore, auto tuning does not run continuously during operation
(this would be adaptive control). Whenever there is a substantial change in loop dynamics, such
as mass of process, size of actuator, etc., the tuning process will need to be repeated in order to
derive new gains required for optimal control.
WARNING: Only authorized personnel fully familiar with all aspects of the process
should make changes that affect the loop tuning constants. Using the loop auto tuning
procedures will affect the process, including inducing large changes in the control output
value. Make sure you thoroughly consider the impact of any changes to minimize the
risk of injury to personnel or damage to equipment. The auto tune in the D4-454 is not
intended to be used as a replacement for your process knowledge.
Once the physical loop components are connected to the PLC, auto tuning can be initiated
within DirectSOFT (see the DirectSOFT Programming Software Manual), and it can be used
to establish initial PID parameter values. Auto tuning is the best “guess” the CPU can do after
some trial tests.
The loop controller offers both closed-loop and open-loop methods. The following sections
describe how to use the auto tuning feature, and what occurs in open and closed-loop auto
tuning.
The controls for the auto tuning function uses bits 5, 6, and 7 in the PID Mode 2 word V+01,
as shown below. DirectSOFT will manipulate these bits automatically when you use the auto
tune feature within DirectSOFT. Or, you may have your ladder logic access these bits directly
for allowing control from another source such as a dedicated operator interface. The individual
control bits allow you to start the auto tune procedure, select PID or PI tuning and select
closed-loop or open-loop tuning. If you select PI tuning, the auto tune procedure leaves the
derivative gain at 0. The Loop Mode and Alarm Status word V+06 reports the auto tune status
as shown. Bit 12 will be on (1) during the auto tune cycle, automatically returning to off (0)
when done.
Auto Tune Function
Start Auto Tune Auto Tune

(0 to 1 transition) Active
0=PID tuning, Auto Tune
Auto Tuning 1=open PI tuning Error Auto Tuning
Controls 0=closed loop, Status
1=open loop
PID Mode 2 Setting V+01 Loop Mode and Alarm Status V+06
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

Open-Loop Auto Tuning

During an open-loop auto tuning cycle, the loop controller operates as shown in the diagram
below. Before starting this procedure, place the loop in Manual Mode and ensure the PV and
control output values are in the middle of their ranges (away from the end points).
PLC System
Process Variable
Response Step Function

Open Loop
Auto Tuning
Control
Setpoint Value Error Term Loop Output Manufacturing
–
Process Variable
NOTE: In theory, the SP value does not matter in this case, because the loop is not closed. However, the
requirement of the firmware is that the SP value must be more than 5% of the PV range from the actual PV
before starting the auto tune cycle (for the D4-454, 12 bit PV should be 205 counts or more below the SP for
forward-acting loops, or 205 counts or more above the SP for reverse-acting loops).
When auto tuning, the loop controller induces a step change on the output and simply observes
the response of the PV. From the PV response, the auto tune function calculates the gains and
the sample time. It automatically places the results in the corresponding registers in the loop
table.
The following timing diagram shows the events which occur in the open-loop auto tuning
cycle. The auto tune function takes control of the control output and induces a 10%-of-span
step change. If the PV change which the loop controller observes is less than 2%, then the step
change on the output is increased to 20%-of-span.
• When Auto Tune starts, step change output m=10%
• During Auto Tune, the controller output reached the full scale positive limit. Auto Tune stopped and
the Auto Tune Error bit in the Alarm word bit turned on.
• When PV change is under 2%, output is changed at 20%. Open Loop Auto Tune Cycle Wave: Step
Response Method.
PV
Tangent Rr = Slope
(%)
SP
Process Wave
Base Line
LrRr
(%)
Lr
(sec.)
Time (sec)
Step Change m=10%
Output Value
(%)
Auto Tune Cycle

PID Cycle PID Cycle
Auto Tune Start Auto Tune End

When the loop tuning observations are complete, the loop controller computes Rr (maximum
slope in %/sec.) and Lr (dead time in sec). The auto tune function computes the gains according
to the Zeigler-Nichols equations, shown below:
PID Tuning PI Tuning
P=1.2*im/LrRr P=0.9*im/LrRr
I=2.0* Lr I=3.33* Lr
D=0.5* Lr D=0
Sample Rate = 0.056* Lr Sample Rate = 0.12*Lr
im = Output step change (10% = 0.1, 20% = 0.2)
We highly recommend using DirectSOFT for the auto tuning interface. The duration of each
auto tuning cycle will depend on the mass of the process. A slowly-changing PV will result in a
longer auto tune cycle time. When the auto tuning is complete, the proportional, integral, and
derivative gain values are automatically updated in loop table locations V+10, V+11, and V+12
respectively. The sample time in V+07 is also updated automatically. You can test the validity
of the values the auto tuning procedure yields by measuring the closed-loop response of the
PV to a step change in the output. The instructions on how to do this are in the section on the
manual tuning procedure (located prior to this auto tuning section).
Closed-Loop Auto Tuning
During a closed-loop auto tuning cycle the loop controller operates as shown in the diagram
below.
PLC System
Process Variable
Response Limit cycle wave

Closed Loop
Auto Tuning
Control
Setpoint V alue Error T erm Loop Output Manufacturing
–
Process Variable
When auto tuning, the loop controller imposes a square wave on the output. Each transition of
the output occurs when the PV value crosses over/under the SP value. Therefore, the frequency
of the limit cycle is roughly proportional to the mass of the process. From the PV response, the
auto tune function calculates the gains and the sample time. It automatically places the results
in the corresponding registers in the loop table.

The following timing diagram shows the events which occur in the closed-loop auto tuning
cycle. The auto tune function examines the direction of the offset of the PV from the SP. The
auto tune function then takes control of the control output and induces a full-span step change
in the opposite direction. Each time the sign of the error (SP – PV) changes, the output changes
full-span in the opposite direction. This proceeds through three full cycles.
This example is direct–acting.
• Mmax = Output Value upper limit setting.
• Mmin = Output Value lower limit setting.
Xo
SP
Process Wave
PV
Output Value
To
PID Cycle PID Cycle
Auto Tune Cycle
Auto Tune Start Auto Tune End

Calculation of
PID parameter
When set to reverse–acting, the output will be inverted. When the loop tuning observations are
complete, the loop controller computes To (bump period) and Xo (amplitude of the PV). Then
it uses these values to compute Kpc (sensitive limit) and Tpc (period limit). From these values,
the loop controller auto tune function computes the PID gains and the sample rate according
to the Zeigler-Nichols equations shown below:
Kpc = 4M / (f *X0) Tpc = 0
M = Amplitude of output
PID Tuning PI Tuning
P = 0.45*Kpc P = 0.30*Kpc
I = 0.60*Tpc I = 1.00*Tpc
D = 0.10*Tpc D=0
Sample Rate = 0.014*Tpc Sample Rate = 0.03*Tpc
Auto tuning error

In open-loop tuning, if the auto tune error bit (bit 13 of loop Mode/Alarm status word V+06)
is on, please verify the PV and SP values are within 5% of full scale difference, as required by
the auto tune function.
NOTE: If your PV fluctuates rapidly, you probably need to use the built-in analog filter (see page 8–55) or
create a filter in ladder logic (see example on page 8–56).

Use DirectSOFT Data View with PID View

The Data View window is a very useful tool which can be used to help tune your PID loop.
You can compare the variables in the PID View with the actual values in the V-memory
location with Data View.
Open a New Data View Window

A new Data View window can be opened in any one of three ways; the menu bar Debug > Data
View > New, the keyboard shortcut Ctrl + Shift + F3 or the Data button on the Status toolbar.
By default, the Data View window is assigned Data1 as the default name. This name can be
changed for the current view using the Options dialog. The following diagram is an example of
a newly opened Data View. The window will open next to the Ladder View by default.
The Data View window can be used just as it is shown above for troubleshooting your PID
logic, and it can be most useful when tuning the PID loop.

Open PID View

The PID View can only be opened after a loop has been setup in your ladder program. PID
View is opened by selecting it from the View submenu on the Menu bar, View > PID View.
The PID View can also be opened by clicking on the PID View button from the PLC Setup
toolbar if it is in view.
The PID View will open and appear over the Ladder View which can be brought into view by
clicking on its tab. When using the Data View and the PID View together, each view can be
sized for better use as shown on the facing page.

The two views are now ready to be used to tune your loop. You will be able to see where the
PID values have been set and see the process that it is controlling.
The diagram below illustrates how to use the views to see the current SP, PV and Output
values, along with the other PID addresses. Refer to the Loop Table Definitions page 8-21 for
details of each word in the table. This is also a good data type reference for each word in the
table.
Scale the time axis of the viewing The trend can be cleared and
window by using this input box. restarted from the left at anytime.
Process Variable
and Setpoint trends are
color coded.
The loop name area

turns red whenever there
is an overflow error.
When both windows are positioned in this manner, you are able to see where the PID values
have been set and see the process that it is controlling.

Using the Special PID Features

It’s a good idea to understand the special features of the D4-454 and how to use them. You may
want to incorporate some of these features for your PID.
How to Change Loop Modes

The first three bits of the PID Mode 1 word (V+00) request
the operating mode of the corresponding loop. Note: these Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
bits are mode change requests, not commands (certain
conditions can prohibit a particular mode change – see next Cascade Manual
Automatic
page).
The normal state of these mode request bits is “000.” To request a mode change, you must
SET the corresponding bit to a “1” using a one-shot. The PID loop controller automatically
resets the bits back to “000” after it reads the mode change request. Methods of requesting
mode changes are:
• DirectSOFT’s PID View – this is the easiest method. Use the pull-down menu, or click
on one of the radio buttons if using older DirectSOFT versions, and the appropriate bit
will get set.
• Ladder program– ladder logic can request any loop mode when the PLC is in Run Mode.
This will be necessary after application startup if mode changes are part of the application.
Use the program shown to the right to SET the mode bit (do Go to Auto Mode
not use an OUT coil). On a 0–1 transition of X0, the rung X0 B2000.1
sets the Auto bit equal to 1. The loop controller resets it. SET
• Operator panel – interface the operator’s panel to

ladder logic using standard methods, then use the logic above to set the mode bit.
Since mode changes can only be requested, the PID loop controller will decide when to permit
mode changes and provide the loop mode status. It reports the current mode on bits 0, 1, and
2 of the Loop Mode/Alarm Status word, location V+06 in the loop table. The parallel request/
monitoring functions are shown in the figure below. The figure also shows the two possible
mode-dependent SP sources, and the two possible Control Output sources.
Manual
Input from Operator
Control Output
from another loop Cascade Control Output
Setpoint Error T erm Loop
+ Calculation
Normal Source – Auto/Cascade
Auto/Manual Process Variable
Mode Select
PID Mode
Control
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Mode Request Mode Monitoring

Cascade Manual Cascade Manual
Automatic Automatic

Operator Panel Control of PID Modes

Since the modes Manual, Auto and Cascade are the most fundamental and important PID
loop controls, you may want to hard-wire mode control switches to an operator’s panel. Most
applications will need only Manual and Auto selections (Cascade is used in special applications).
Remember that mode controls are really mode request bits, and the actual loop mode is indicated
elsewhere.
The following figure shows an operator’s panel using momentary push-buttons to request PID
mode changes. The panel’s mode indicators do not connect to the switches, but interface to the
corresponding data locations.
Operator’s Panel
Manual
Auto
Mode Request Mode Monitoring
Cascade
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
PLC Modes Effect on Loop Modes

If you have selected the option for the loops to follow the PLC mode, the PLC modes (Program,
Run) interact with the loops as a group. The following summarizes this interaction:
• When the PLC is in Program Mode, all loops are placed in Manual Mode and no loop
calculations occur. However, note that output modules (including analog outputs) turn
off in PLC Program Mode. So, actual manual control is not possible when the PLC is in
Program Mode.
• The only time the CPU will allow a loop mode change is during PLC Run Mode operation.
As such, the CPU records the modes of all 16 loops as the desired mode of operation. If
power failure and restoration occurs during PLC Run Mode, the CPU returns all loops to
their prior mode (which could be Manual, Auto, or Cascade).
• On a Program-to-Run mode transition, the CPU forces each loop to return to its prior
mode recorded during the last PLC Run Mode.
• You can add and configure new loops only when the PLC is in Program Mode. New loops
automatically begin in Manual Mode.

Loop Mode Override

In normal conditions the mode of a loop is determined by the request to V+00, bits 0,
1, and 2. However, some conditions exist which will prevent a requested mode change
from occurring:
• A loop that is not set independent of PLC mode cannot change modes when the PLC is
in Program mode.
• A major loop of a cascaded pair of loops cannot go from Manual to Auto until its minor
loop is in Cascade mode.
In other situations, the PID loop controller will automatically change the mode of the loop to
ensure safe operation:
• A loop which develops an error condition automatically goes to Manual.
• If the minor loop of a cascaded pair of loops leaves Cascade Mode for any reason, its major
loop automatically goes to Manual Mode.

PV Analog Filter
A noisy PV signal can make tuning difficult and can cause the control output to be more
extreme than necessary, as the output tries to respond to the peaks and valleys of the PV. There
are two equivalent methods of filtering the PV input to make the loop more stable. The first
method is accomplished using the D4-454’s built-in filter. The second method achieves a
similar result using ladder logic.
The D4-454 Built-in Analog Filter
The D4-454 provides a selectable first-order low-pass PV input filter. We only recommend
the use of a filter during auto tuning or PID control if there is noise on the input signal. You
may disable the filter after auto tuning is complete, or continue to use it if the PV input signal
is noisy.
Loop Control Output

+ Calculation
– Unfiltered
PV
0
Process Variable
1 Filtered
PV
P ID Mode 2 Setting V+01
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Loop Table

PV filter V+24 XXXX FIlter constant
enable/disable
Bit 2 of PID Mode Setting 2 provides the enable/disable control for the low-pass PV filter
(0=disable, 1=enable). The roll-off frequency of the single-pole low-pass filter is controlled by
using register V+24 in the loop parameter table, the filter constant. The data format of the filter
constant value is BCD, with an implied decimal point 00X.X, as follows:
• The filter constant has a valid range of 000.1 to 001.0. The smaller the filter value, the
greater the filtering performed; for example, the value 001.0 provides no filtering.
• DirectSOFT converts values above the valid range to 001.0 and values below this range
to 000.1
• Values close to 001.0 result in higher roll-off frequencies, while values closer to 000.1
result in lower roll-off frequencies.

The algorithm which the built-in filter follows is:

yi = k (xi – yi–1) + yi–1
yi is the current output of the filter
xi is the current input to the filter
yi–1 is the previous output of the filter
k is the PV Analog Input Filter Factor
Creating an Analog Filter in Ladder Logic

A similar algorithm can be built in your ladder program. Your analog inputs can be filtered
effectively using either method. The following programming example describes the ladder
logic you will need. Be sure to change the example memory locations to those that fit your
application.
Filtering can induce a 1 part in 1000 error in your output because of “rounding.” If your
process cannot tolerate a 1 part in 1000 error, do not use filtering. Because of the rounding
error, you should not use zero or full scale as alarm points. Additionally, the smaller the filter
constant the greater the smoothing effect, but the slower the response time. Be sure a slower
response is acceptable in controlling your process.
.
. SP1
Loads the analog signal, which is a BCD value
LD and has been loaded from V-memory location
V2000
V2000, into the accumulator. Contact SP1 is
always on.
Converts the BCD value in the accumulator
BIN to binary. This instruction is not needed if the
analog value is originally brought in as a
binary number.
BTOR Converts the binary value in the accumulator

to a real number.
Subtracts the real number stored in location

SUBR V1400 from the real number in the
V1400 accumulator, and stores the result in the
accumulator. V1400 is the designated
workspace in this example.
Multiplies the real number in the
MULR accumulator by 0.2 (the filter factor),
R0.2 and stores the result in the
accumulator. This is the filtered value.
Adds the real number stored in
ADDR location V1400 to the real number
V1400 filtered value in the accumulator, and
stores the result in the accumulator.
Copies the value in the accumulator

OUTD to location V1400.
V1400
Converts the real number in the

Maintenance
RTOB accumulator to a binary value, and

stores the result in the accumulator.
Converts the binary value in the accumulator

BCD to a BCD number. Note: the BCD instruction
is not needed for PID loop PV (loop PV is a
binary number).
Loads the BCD number filtered value from

. OUT the accumulator into location V1402 to use
. V1402
in your application or PID loop.

Use the DirectSOFT Filter Intelligent Box Instruction

For those who are using DirectSOFT, you have the opportunity to use Intelligent Box (IBox)
instruction IB-402, Filter Over Time in Binary (decimal). This IBox will perform a first-order
filter on the Raw Data on a defined time interval. The equation is,
New = Old + [(Raw - Old) / FDC] where:
• New =New Filtered Value
• Old = Old Filtered Value
• FDC = Filter Divisor Constant
• Raw = Raw Data
The Filter Divisor Constant is an integer
in the range K1 to K100, such that if it
equaled K1, then no filtering is performed.
The rate at which the calculation is performed is specified by time in hundredths of a second
(0.01 seconds) as the Filter Freq Time parameter. Note that this Timer instruction is embedded
in the IBox and must NOT be used any other place in your program. Power flow controls
whether the calculation is enabled. If it is disabled, the Filter Value is not updated. On the first
scan from Program to Run mode, the Filter Value is initialized to 0 to give the calculation a
consistent starting point.
Since the following binary filter example does not write directly to the PID PV location, use
ladder to move the Filtered Value result to the PID PV location (V+03).
FilterB Example
Following is an example of how the FilterB IBox is used in a ladder program. The instruction
is used to filter a binary value that is in V2000. Timer (T1) is set to 0.5 seconds, the rate at
which the filter calculation will be performed. The filter constant is set to 3.0. A larger value
will increase the smoothing effect of the filter. A value of 1 results with no filtering. The filtered
value will be placed in V2100.
NOTE: For information on the Filter Over Time-Binary Ibox IB-402, see the DL405-IBOX-S manual.

Ramp/Soak Generator
Introduction
Our discussion of basic loop operation noted the setpoint for a loop will be generated in
various ways, depending on the loop operating mode and programming preferences. In the
figure below, the ramp/soak generator is one of the ways the SP may be generated. It is the
responsibility of your ladder program to ensure only one source attempts to write the SP value
at V+02 at any particular time.
Setpoint Sources:
Operator Input
Setpoint V+02 Loop Control Output
Ramp/soak generator
+ Calculation
Ladder Program –
Another loop’s output (cascade)
Process Variable
If the SP for your process rarely changes or can tolerate step changes, you probably will not
need to use the ramp/soak generator. However, some processes require precisely-controlled SP
value changes. The ramp/soak generator can greatly reduce the amount of programming required
for these applications.
The terms ramp and soak have special meanings in the
process control industry, and refer to desired setpoint (SP) SP
values in temperature control applications. In the figure to Soak
the right, the setpoint increases during the ramp segment. Ramp
It remains steady at one value during the soak segment.
slope
Complex SP profiles can be generated by specifying a
series of ramp/soak segments. The ramp segments are Time
specified in SP units per second time. The soak time is
also programmable in minutes.
It is instructive to view the ramp/soak generator as a dedicated function to generate SP values,
as shown below. It has two categories of inputs which determine the SP values generated. The
ramp/soak table must be programmed in advance, containing the values that will define the
ramp/soak profile. The loop reads from the table during each PID calculation as necessary. The
ramp/soak controls are bits in a special loop table word that control the real-time start/stop
functionality of the ramp/soak generator. The ladder program can monitor the status of the
ramp soak profile (current ramp/segment number).
Ramp/soak table
Ramp/soak Setpoint Loop Control Output
Ramp/soak controls Generator + Calculation
–
Process Variable

Now that we have described the general ramp/soak generator operation, we list its specific
features:
• Each loop has its own ramp/soak generator (use is optional).
• You may specify up to eight ramp/soak steps (16 segments).
• The ramp soak generator can run any time the PLC is in Run mode. Its operation is independent of
the loop mode (Manual or Auto).
• Ramp/soak real-time controls include Start, Hold, Resume, and Jog.
• Ramp/soak monitoring includes Profile Complete, Soak Deviation (SP minus PV), and current
ramp/soak step number.
The following figure shows an SP profile consisting of ramp/soak segment pairs. The segments
are individually numbered as steps from 1 to 16. The slope of each of the ramp segments may
be either increasing or decreasing. The ramp/soak generator automatically knows whether to
increase or decrease the SP based on the relative values of a ramp’s end points. These values
come from the ramp/soak table.
15 16
13 14 Soak
Ramp
5 6 Soak
Ramp
3 4 Soak
Ramp
Step 1 2 Soak
Ramp
Soak
Ramp
SP
Ramp/Soak Table
The parameters which define the ramp/soak profile for V–Memory Space
a loop are in a ramp/soak table. Each loop may have its
own ramp/soak table, but it is optional. Recall the Loop User Data
Parameter table consists of a 32-word block of memory V2000 LOOP #1 V2034 =

for each loop, and together they occupy one contiguous V2037 32 words 3000 octal
memory area. However, the ramp/soak table for a loop V2040 LOOP #2
is individually located, because it is optional for each 32 words V2074 =
V2077
3600 octal
loop. An address pointer in location V+34 in the loop
table specifies the starting location of the ramp/soak
V3000
table. Ramp/Soak #1
32 words
In the example to the right, the loop parameter tables
for Loop #1 and #2 occupy contiguous 32-word blocks
as shown. Each has a pointer to its ramp/soak table,
independently located elsewhere in user V-memory. Of
V3600 Ramp/Soak #2
course, you may locate all the tables in one group, as 32 words
long as they do not overlap.

The parameters in the ramp/soak table must be user-defined. the most convenient way is to
use DirectSOFT, which features a special editor for this table. Four parameters are required to
define a ramp and soak segment pair, as pictured below.
• Ramp End Value – specifies the destination SP value for the end of the ramp. Use the same data
format for this number as you use for the SP. It may be above or below the beginning SP value, so
the slope could be up or down (we don’t have to know the starting SP value for ramp #1).
• Ramp Slope – specifies the SP increase in counts (units) per second. It is a BCD number from 00.00
to 99.99 (uses implied decimal point).
• Soak Duration – specifies the time for the soak segment in minutes, ranging from 000.1 to 999.9
minutes in BCD (implied decimal point).
• Soak PV Deviation – (optional) specifies an allowable PV deviation above and below the SP value
during the soak period. A PV deviation alarm status bit is generated by the ramp/soak generator.
Ramp End
SP Value Soak PV Ramp/Soak Table
deviation V+00 XXXX Ramp End SP Value
V+01 XXXX Ramp Slope
Slope Soak
SP V+02 XXXX Soak Duration
duration
V+03 XXXX Soak PV Deviation
segment becomes active
The ramp segment becomes active when the previous soak segment ends. If the ramp is the first
segment, it becomes active when the ramp/soak generator is started, and automatically assumes
the present SP as the starting SP.
Offset Step Description Offset Step Description


Many applications do not require all 16 R/S steps. Use all zeros in the table for unused steps.
The R/S generator ends the profile when it finds ramp slope = 0.
Ramp/Soak Table Flags

The individual bit definitions of the Ramp/Soak Table Flag (Addr+33) word is listed in the
following table.
Bit Ramp/Soak Flag Bit Description Read/Write Bit=0 Bit=1
0 Start Ramp / Soak Profile Write – 0 –> 1 Start
1 Hold Ramp / Soak Profile Write – 0 –> 1 Hold
2 Resume Ramp / Soak Profile Write – 0 –> 1 Resume
3 Jog Ramp / Soak Profile Write – 0 –> 1 Jog
4 Ramp / Soak Profile Complete Read – Complete
5 PV Input Ramp / Soak Deviation Read Off On
6 Ramp / Soak Profile in Hold Read Off On
7 Reserved Read Off On
8–15 Current Step in R/S Profile Read Decode as byte (hex)
Ramp/Soak Generator Enable

The main enable control to permit ramp/soak generation
of the SP value is accomplished with bit 11 in the PID
Mode 1 Setting V+00 word, as shown to the right. The Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
other ramp/soak controls in V+33 shown in the table Ramp/Soak

above will not operate unless this bit=1 during the entire Generator Enable
ramp/soak process.
Ramp/Soak Controls
The four main controls for the ramp/soak generator are
in bits 0 to 3 of the Ramp/Soak Setting V + 33 word Ramp/Soak Settings V+33
in the loop parameter table. DirectSOFT controls these
bits directly from the ramp/soak settings dialog. However, Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
you must use ladder logic to control these bits during
Jog
program execution. We recommend using the bit-of-
Resume
word instructions.
Hold
Ladder logic must set a control bit to a 1 to command the Start
corresponding function. When the loop controller reads
the ramp/soak value, it automatically turns off the bit for
you. Therefore, a reset of the bit is not required, when the
CPU is in Run Mode.
The example program rung to the right shows how an Start R/S Generator
external switch X0 can turn on, and the PD contact uses X0 B2033.0
the leading edge to set the proper control bit to start SET
the ramp soak profile. This uses the Set Bit-of-word
instruction.

The normal state for the ramp/soak control bits is all zeros. Ladder logic must set only one
control bit at a time.
• Start – a 0 to 1 transition will start the ramp soak profile. The CPU must be in Run Mode, and the
loop can be in Manual or Auto Mode. If the profile is not interrupted by a Hold or Jog command,
it finishes normally.
• Hold – a 0 to 1 transition will stop the ramp/soak profile in its current state, and the SP value will
be frozen.
• Resume – a 0 to 1 transition will cause the ramp/soak generator to resume operation if it is in the
hold state. The SP values will resume from their previous value.
• Jog – a 0 to 1 transition will cause the ramp/soak generator to truncate the current segment (step),
and go to the next segment.
Ramp/Soak Profile Monitoring

You can monitor the Ramp/Soak profile status using other bits Ramp/Soak Settings V+33
in the Ramp/Soak Settings V+33 word, shown to the right. Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
R/S Profile in Hold

• R/S Profile Complete =1 when the last programmed step is done. Soak PV Deviation
R/S Profile Complete
• Soak PV Deviation =1 when the error (SP–PV) exceeds the
specified deviation in the R/S table.
• R/S Profile in Hold =1 when the profile was active but is now
in hold.
• The number of the current step is available in the upper Ramp/Soak Settings V+33
8 bits of the Ramp/Soak Settings V+33 word. The bits Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
represent a 2-digit hex number, ranging from 1 to 10.

Ladder logic can monitor these to synchronize other Current Profile Step, 2–digit hex
parts of the program with the ramp/soak profile. Load Value = 01 to 10 hex,
or 1 to 16 decimal
this word to the accumulator and shift right 8 bits, and
you have the step number.
Ramp/Soak Programming Errors

Ramp/Soak Table Error V+35
The starting address for the ramp/soak table must be a valid
location. If the address points outside the range of user Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
V-memory, one of the bits to the right will turn on when Starting Address set in
reserved system V-memory
the ramp/soak generator is started. We recommend using
Starting Address set out of
DirectSOFT to configure the ramp/soak table. It automatically V-memory upper range
range checks the addresses for you. Starting Address set out
of V-memory lower range
Testing Your Ramp/Soak Profile
It’s a good idea to test your ramp/soak profile before using it to control the process. This is easy
to do, because the ramp/soak generator will run even when the loop is in Manual Mode. Using
DirectSOFT’s PID View will be a real time-saver, because it will draw the profile on-screen for
you. Be sure to set the trending timebase slow enough to display completed ramp-soak segment
pairs in the waveform window.

DirectSOFT Ramp/Soak Example

The following example will step you through the Ramp/Soak setup.
Setup the Profile in PID Setup

The first step is to use Setup PID in DirectSOFT to set the profile of your process. Open the
Setup PID window and select the R/S tab, and then enter the Ramp and Soak data.
Program the Ramp/Soak Control in Relay Ladder

Refer to the Ramp/Soak Flag Bit Description table on page 8-63 when adding the control
rungs to your program similar to the ladder rungs below.

Test the Profile

After the Ramp/Soak program has been developed in RLL, test the program. Check your
profile by using PID View. If there are any changes to be made in the profile, they are made
in the PID Setup R/S profile. Make the changes in Program mode then start the Ramp/Soak
process again.

Cascade Control
Introduction
Using cascaded loops is an advanced control technique, superior to individual loop control in
certain situations. As the name implies, cascade means that one loop is connected to another
loop. In addition to Manual (open loop) and Auto (closed loop) Modes, the D4-454 also
provides Cascaded Mode.
NOTE: Using cascaded loops is an advanced process control technique; therefore, we recommend their use
only for experienced process control engineers.
When a manufacturing process is complex and contains a lag time from control input to
process variable output, even the most perfectly tuned single loop around the process may yield
slow and inaccurate control. It may be the actuator operates on one physical property, which
eventually affects the process variable, measured by a different physical property. Identifying the
intermediate variable allows us to divide the process into two parts as shown in the following
figure.
PROCESS
Intermediate Process
Control input Process A Variable Process B Variable (PV)
The principle of cascaded loops is simply that we add another process loop to more precisely control
the intermediate variable! This separates the source of the control lag into two parts, as well.
The diagram below shows a cascade control system, showing that it is simply one loop nested
inside another. The inside loop is called the minor loop, and the outside loop is called the
major loop. For overall stability, the minor loop must be the fastest responding loop of the
two (try a factor of 10 for a better response time). We do have to add the additional sensor to
measure the intermediate variable (PV for process A). Notice the setpoint for the minor loop
is automatically generated for us, by using the output of the major loop. Once the cascaded
control is programmed and debugged, we only need to deal with the original setpoint and
process variable at the system level. The cascaded loops behave as one loop, but with improved
performance over the previous single-loop solution.
External External
Disturbances Disturbances
Output B/
Setpoint Loop B Setpoint A Loop A Output A Process A Process B
+ Calculation + Calculation (secondary) (primary)
– –
Minor
Major Loop
Loop PV, Process A
PV, Process B
One of the benefits to cascade control can be seen by examining its response to external
disturbances. Remember the minor loop is faster acting than the major loop. Therefore, if a
disturbance affects process A in the minor loop, the Loop A PID calculation can correct the
resulting error before the major loop sees the effect.

Cascaded Loops in the D4-454 CPU

In the use of the term cascaded loops, we must make an important distinction. Only the minor
loop will actually be in the Cascade Mode. In normal operation, the major loop must be in
Auto Mode. If you have more than two loops cascaded together, the outer-most (major) loop
must be in Auto Mode during normal operation, and all inner loops in Cascade Mode.
NOTE: Technically, both major and minor loops are cascaded in strict process control terminology.
Unfortunately, we are unable to retain this convention when controlling loop modes. Remember that all minor
loops will be in Cascade Mode, and only the outer-most (major) loop will be in Auto Mode.
You can cascade together as many loops as necessary on the D4-454, and you may have multiple
groups of cascaded loops. For proper operation on cascaded loops you must use the same data
range (12/15 bit) and unipolar/bipolar settings on the major and minor loop.
To prepare a loop for Cascade Mode operation as a minor loop, you must program its remote
Setpoint Pointer in its loop parameter table location V+32, as shown below. The pointer must
be the address of the V+05 location (control output) of the major loop. In Cascade Mode, the
minor loop will ignore its local SP register (V+02), and read the major loop’s control output
as its SP instead.
Major Loop (Auto mode) Minor Loop (Cascade Mode)
Loop Table Loop Table
V+02 XXXX SP V+02 XXXX SP
V+03 XXXX PV V+03 XXXX PV
V+05 XXXX Control Output V+05 XXXX Control Output
V+32 XXXX Remote SP Pointer
When using DirectSOFT’s PID View to watch the SP value of the minor loop, DirectSOFT
automatically reads the major loop’s control output and displays it for the minor loop’s SP. The
minor loop’s normal SP location, V+02, remains unchanged.
Now, we use the loop parameter arrangement above and draw its equivalent loop diagram,
shown below.
Major loop Minor Cascaded loop
Control Output V+05 Cascade

Loop
Control
Calculation
Remote Setpoint Loop Output
SP + Calculation
Local SP
–
V+02 Auto/Manual Process Variable
Remember that a major loop goes to Manual Mode automatically if its minor loop is taken out
of Cascade Mode.

Tuning Cascaded Loops

In tuning cascaded loops, you will need to de-couple the cascade relationship and tune the
loops individually, using one of the loop tuning procedures previously covered.
1. If you are not using auto tuning, then find the loop sample rate for the minor loop, using the
method discussed earlier in this chapter. Then set the sample rate of the major loop slower than the
minor loop by a factor of 10. Use this as a starting point.
2. Tune the minor loop first. Leave the major loop in Manual Mode, and you will need to generate SP
changes for the minor loop manually as described in the loop tuning procedure.
3. Verify the minor loop gives a critically-damped response to a 10% SP change while in Auto Mode.
Then we are finished tuning the minor loop.
4. In this step, you will need to get the minor loop in Cascade Mode, and then the Major loop in
Auto Mode. We will be tuning the major loop with the minor loop treated as a series component
in its overall process. Therefore, do not go back and tune the minor loop again while tuning the
major loop.
5. Tune the major loop, following the standard loop tuning procedure in this section. The response of
the major loop PV is actually the overall response of the cascaded loops together.

Time-Proportioning Control
The PID loop controller in the D4-454 CPU generates a smooth control output signal across a
numerical range. The control output value is suitable to drive an analog output module, which
connects to the process. In the process control field, this is called continuous control, because the
output is on (at some level) continuously.
While continuous control can be smooth and robust, the cost of the loop components (such
as actuator, heater amplifiers) can be expensive. A simpler form of control is called time-
proportioning control. This method uses actuators which are either on or off (no in-between).
Loop components for on/off-based control systems are lower cost than their continuous control
counterparts.
In this section, we will show you how to convert the control output of a loop to time-
proportioning control for the applications that need it. Let’s take a moment to review how
alternately turning a load on and off can control a process. The diagram below shows a hot-air
balloon following a path across some mountains. The desired path is the setpoint. The balloon
pilot turns the burner on and off alternately, which is his control output. The large mass of air
in the balloon effectively averages the effect of the burner, converting the bursts of heat into a
continuous effect: slowly changing balloon temperature and ultimately the altitude, which is
the process variable.
Time-proportioning control approximates continuous control by virtue of its duty-cycle the

ratio of ON time to OFF time. The following figure shows an example of how duty-cycle
approximates a continuous level when it is averaged by a large process mass.
period
Desired
Effect
On/Off On
Control Off
If we were to plot the on/off times of the burner in the hot-air balloon, we would probably see
a very similar relationship to its effect on balloon temperature and altitude.

On/Off Control Program Example

The following ladder segment provides a time proportioned on/off control output. It converts
the continuous output in V2005 to on/off control using the output coil, Y0.
SP Loop V2005 Time Y0 P

Process
+ Calculation Proportioning V
–
continuous on/off
PV
The example program uses two timers to generate On/Off control. It makes the following
assumptions, which you can alter to fit your application:
• The loop table starts at V2000, so the control output is at V2005.
• The data format of the control output is 12-bit, unipolar (0 – FFF or 0 – 4095).
• The time base (one full cycle) for the On/Off waveform is 10 seconds. We
use a fast timer (0.01 sec/tick), counting to 1000 ticks (10 seconds).
• The On/Off control output is Y0.
The time proportioning program must match the resolution of the output (1 part in 1000) to
the resolution of the time base of T0 (also 1 part in 1000).
NOTE: Some processes change too fast for time proportioning control. Consider the speed of your process
when you choose this control method. Use continuous control for processes that change too fast for time
proportioning control. Also, consider using a solid state switch for a longer switch life instead of a relay.
T0 A fast timer (0.01 sec. timebase) establishes the primary

TMRF T0 time interval. The constant, K1000, sets the preset at 10
K1000 seconds (1,000 ticks). The N.C. enabling contact, T0,
makes the timer self-resetting. T0 is on for one scan
each 10 seconds, when it resets itself and T1.
T0 At the end of the 10 second period, T0 turns on, and
LD loads the control output value (binary) from the loop table
V2005
V+05 location (V2005).
BTOR The BTOR instruction changes the number in the

accumulator to a real number.
Dividing the control output by 4.095, converts the

DIVR 0 – 4095 range to 0 – 1000, which “matchs” the
R4.095 number of ticks in the 10 second timer range.
RTOB This instruction converts the real number back to

binary. This step prepares the number for conversion
to BCD. There is no real-to-BCD instruction.
BCD Convert the number in the accumulator to BCD format.
This satisfies the timer preset format requirement.
Output the result to V1400. In our example, this is the

OUT location of the timer preset for the second timer.
V1400
T0 The second fast timer also counts in increments of .01

TMRF T1 seconds, so its range is variable from 0 to a maximum
V1400 of 1000 ticks, or 10 seconds. This timer’s output, T1,
turns off the output coil, Y0, when the preset is reached.
T1 TA1 K0 The N.C. T1 contact, inverts the T1 timer output. The

Y0
control output is on at the beginning of the 10-second time
OUT interval. Y0 turns off when T1 times out. The STRNE
contact prevents Y0 from energizing during the one scan
when T0 resets T1. Y0 is the actual control output.
END END coil marks the end of the main program.

Feedforward Control
Feedforward control is an enhancement to standard closed-loop control. It is most useful for
diminishing the effects of a quantifiable and predictable loop disturbance or sudden change in
setpoint. Use of this feature is an option available to you on the D4-454. However, it’s best to
implement and tune a loop without feedforward, and adding it only if better loop performance
is still needed. The term feedforward refers to the control technique involved, shown in the
diagram below. The incoming setpoint value is fed forward around the PID equation, and
summed with the output.
Feedforward path
kf
+
Setpoint Loop Control Output
+ Calculation +
–
Process Variable
In the previous section on the bias term, we said that “the bias term value establishes a working
region or operating point for the control output. When the error fluctuates around its zero point,
the output fluctuates around the bias value.” Now, when there is a change in setpoint, an error
is generated and the output must change to a new operating point. This also happens if a
disturbance introduces a new offset in the loop. The loop does not really know its way to
the new operating point... the integrator (bias) must increment/decrement until the error
disappears, and then the bias has found the new operating point.
Suppose that we are able to know a sudden setpoint change is about to occur (common in some
applications). We can avoid much of the resulting error in the first place, if we can quickly
change the output to the new operating point. If we know (from previous testing) what the
operating point (bias value) will be after the setpoint change, we can artificially change the
output directly (which is feedforward). The benefits of using feedforward are:
• The SP–PV error is reduced during predictable setpoint changes or loop offset disturbances.
• Proper use of feedforward will allow us to reduce the integrator gain. Reducing integrator
gain gives us an even more stable control system.
Feedforward is very easy to use in the D4-454 loop controller, as shown below. The bias term
has been made available to the user in a special read/write location, at PID Parameter Table
location V+04.
Parameter Table location V+04.
Loop Calculation
kp P
V+04
Setpoint Error T erm + Control Output
+ ki I XXXX Bias Term
+
– +
Process Variable kd D

To change the bias (operating point), ladder logic only has to write the desired value to V+04.
The PID loop calculation first reads the bias value from V+04 and modifies the value based
on the current integrator calculation. Then it writes the result back to location V+04. This
arrangement creates a sort of transparent bias term. All you have to do to implement feed
forward control is write the correct value to the bias term at the right time (see the following
example).
NOTE: When writing the bias term, one must be careful to design ladder logic to write the value only once,
at the moment when the new bias operating point is to occur. If ladder logic writes the bias value on every
scan, the loop’s integrator is effectively disabled.
Feedforward Example
How do we know when to write to the bias term, and what value to write? Suppose we have an
oven temperature control loop, and we have already tuned the loop for optimal performance.
Refer to the figure below. We notice that when the operator opens the oven door, the
temperature sags a bit while the loop bias adjusts to the heat loss. Then, when the door closes,
the temperature rises above the SP until the loop adjusts again. Feedforward control can help
diminish this effect.
Oven Closed
Open Closed
door
PV PV sags
PV excess
Bias
First, we record the amount of bias change the loop controller generates when the door opens
or closes. Then, we write a ladder program to monitor the position of an oven door limit
switch. When the door opens, our ladder program reads the current bias value from V+04, adds
the desired change amount, and writes it back to V+04. When the door closes, we duplicate
the procedure, but subtracting desired change amount instead. The following figure shows the
results.
Oven Closed
Open Closed
door
PV
Feed-forward Feed-forward
Bias
The step changes in the bias are the result of our two feed-forward writes to the bias term. We
can see the PV variations are greatly reduced. The same technique may be applied for changes
in setpoint.

PID Example Program

Program Setup for the PID Loop
After the PID loop(s) has been setup with DirectSOFT, you will need to edit your RLL
program to include the rungs needed to setup the analog I/O module to be used by the PID
loop(s).
The example program shows how an analog input module, F4-08AD is used to setup a PID
loop. This example assumes that the PID table for loop 1 has a beginning address of V3000.
All the analog I/O modules used with the D4-454 are setup in a similar manner. Refer to the
DL405 Analog I/O Manual for the setup information for the module that you will be using.
Setup for a F4-08AD module located in Slot 1 to use eight points.

The inputs will be read in binary format, 0-4095.
For this example, the value for each of the input is stored in V-memory location V2000.
V2007; one input per scan. Input Ch.1 = V2000, Ch. 2 = V2001, Ch. 3 = V2002, etc.
PID loop 1 starts at V3000.
On
SP1 LDF
1
Loop 1 Manual
XO
K12
LDF
X14
K3
OUTX
CH 1 Input
Store the binary value for Ch.1 input into the V-memory for PID loop 1 process value, PV V2000
On
SP1 LD
2
CH 1 Input
V2000
OUT
Loop 1 PID PV
V3003
Store PID loop 1 control value, CV, (binary format) to a V-memory location for use in the
program.
On
SP1 LD
Maintenance
3
Loop 1 PID Output
V3005
OUT
V4000

The setpoint (SP) value stored in V-memory is converted from BCD to binary and stored to
PID loop 1 setpoint, SP.
NOTE: The value stored in V1400 must be in the same scale as the PV value.
On
SP1 LD
4
V1400
BIN
OUT
Loop 1 PID SP
V3002
Select PID Loop 1 Manual Mode
Loop 1 Manual Manual Mode Rqst

X0 B3000.0
5 ( SET )
Select PID Loop 1Auto Mode
Loop 1 Auto Auto Mode Rqst

X1 B3000.1
6 ( SET )
7 ( END )
8 ( NOP )
.

Note that the modules used in the PID loop example program were set up for binary format.
They could have been set up for BCD format. In the later case, the BCD data would have to
be converted to binary format before being stored to the setpoint and process variable, and the
control output would have to be converted from binary to BCD before being stored to the
analog output.
By following the steps outlined in this chapter, you should be able to setup workable PID
control loops. The DirectSOFT Programming Software Manual provides more information
for the use of PID View.
For a step-by-step tutorial, go to the Technical Support section located on our website,
automationdirect.com. Once you are at the website, click on Technical Support Home. After
this page opens, find and select Guided Tutorials located under the Using Your Products
column. An Animated Tutorial page will open. Under Available Tutorials, find PID Trainer
and select View the Powerpoint slide show and begin viewing the tutorial. The Powerpoint
Viewer can be downloaded if your computer does not have Powerpoint installed.

Troubleshooting Tips
Q. The loop will not go into Automatic Mode.
A. Check the following for possible causes:
• A PV alarm exists, or a PV alarm programming error exists.
• The loop is the major loop of a cascaded pair, and the minor loop is not in Cascade Mode.
Q. The Control Output stays at zero constantly when the loop is in Automatic
Mode.
• The Control Output upper limit in loop table location V+31 is zero.
• The loop is driven into saturation, because the error never goes to zero value and changes
(algebraic) sign.
Q. The Control Output value is not zero, but it is incorrect.

• The gain values are entered improperly. Remember, gains are entered in the loop table in
BCD, while the SP and PV are in binary. If you are using DirectSOFT, it displays the SP,
PV, Bias and Control output in decimal (BCD), converting it to binary before updating
the loop table.
Q. The Ramp/Soak Generator does not operate when I activate the Start bit.
• The Ramp/Soak enable bit is off. Check the status of bit 11 of loop parameter table
location V+00. It must be set =1.
• The hold bit or other bits in the Ramp/Soak control are on.
• The beginning SP value and the first ramp ending SP value are the same, so first ramp
segment has no slope and consequently has no duration. The ramp/soak generator moves
quickly to the soak segment, giving the illusion the first ramp is not working.
• The loop is in Cascade Mode, and is trying to get the SP remotely.
• The SP upper limit value in the loop table location V+27 is too low.
• Check your ladder program to verify it is not writing to the SP location (V+02 in the loop
table). A quick way to do this is to temporarily place an end coil at the beginning of your
program, then go to PLC Run Mode, and manually start the ramp/soak generator.

Q. The PV value in the table is constant, even though the analog module receives
the PV signal.
A. Your ladder program must read the analog value from the module successfully and write it
into the loop table V+03 location. Verify the analog module is generating the value, and the
ladder is working.
Q. The Derivative gain doesn’t seem to have any affect on the output.
A. The derivative limit is probably enabled (see section on derivative gain limiting).
Q. The loop Setpoint appears to be changing by itself.

• The Ramp/Soak generator is enabled, and is generating setpoints.
• If this symptom occurs on loop Manual-to-Auto Mode changes, the loop is in Bumpless
Transfer 1.
• Check your ladder program to verify it is not writing to the SP location (V+02 in the loop
table). A quick way to do this is to temporarily place an end coil at the beginning of your
program, then go to PLC Run Mode.
Q. The SP and PV values I enter with DirectSOFT work okay, but these values
do not work properly when the ladder program writes the data.
A. The PID View in DirectSOFT lets you enter SP, PV, and Bias values in decimal, and
displays them in decimal for your convenience. For example, when the data format is 12 bit
unipolar, the values range from 0 to 4095. However, the loop table actually requires these in
hex, so DirectSOFT converts them for you. The values in the table range from 0 to FFF, for
12-bit unipolar format. Your ladder program must convert constant values from their BCD
format (when entered as Kxxxx) to binary with the BIN instruction or you must enter them in
the constant field (Kxxxx) as the hex equivalent of the decimal value.
Q. The loop seems unstable and impossible to tune, no matter what gains I use.
• The loop sample time is set too long. Refer to the section near the front of this chapter on
selecting the loop update time.
• The gains are too high. Start out by reducing the derivative gain to zero. Then reduce the
integral gain, and the proportional gain if necessary.
• There is too much transfer lag in your process. This means the PV reacts sluggishly to
control output changes. There may be too much “distance” between actuator and PV
sensor, or the actuator may be weak in its ability to transfer energy into the process.
• There may be a process disturbance that is over-powering the loop. Make sure the PV is
relatively steady when the SP is not changing.

Glossary of PID Loop Terminology

Automatic Mode An operational mode of a loop, in which it makes PID calculations and updates the
loop’s control output.
Bias Freeze A method of preserving the bias value (operating point) for a control output, by inhibiting the
integrator when the output goes out-of-range. The benefit is a faster loop recovery.
Bias Term In the position form of the PID equation, it is the sum of the integrator and the initial control
output value.
Bumpless Transfer A method of changing the operation mode of a loop while avoiding the usual sudden
change in control output level. This consequence is avoided by artificially making the SP and PV equal,
or the bias term and control output equal at the moment of mode change.
Cascaded Loops A cascaded loop receives its setpoint from the output of another loop. Cascaded loops
have a major/minor relationship, and work together to ultimately control one PV.
Cascade Mode An operational mode of a loop, in which it receives its SP from another loop’s output.
Continuous Control Process control done by delivering a smooth (analog) signal as the control output.
Control Output The numerical result of a PID equation which is sent by the loop with the intention of
nulling out the current error.
Derivative Gain A constant that determines the magnitude of the PID derivative term in response to the
current error.
Direct-Acting Loop A loop in which the PV increases in response to a control output increase. In other
words, the process has a positive gain.
Error The difference in value between the SP and PV, Error = SP – PV
Error Deadband An optional feature which makes the loop insensitive to errors when they are small. You
can specify the size of the deadband.
Error Squared An optional feature which multiplies the error by itself, but retains the original algebraic
sign. It reduces the effect of small errors, while magnifying the effect of large errors.
Feedforward A method of optimizing the control response of a loop when a change in setpoint or
disturbance offset is known and has a quantifiable effect on the bias term.
Integral Gain A constant that determines the magnitude of the PID integral term in response to the
current error.
Major Loop In cascade control, it is the loop that generates a setpoint for the cascaded loop.
Manual Mode An operational mode of a loop, in which the PID calculations are stopped. The operator
must manually control the loop by writing to the control output value directly.
Minor Loop In cascade control, the minor loop is the subordinate loop that receives its SP from the major
loop.
On/Off Control A simple method of controlling a process, through on/off application of energy into the
system. The mass of the process averages the on/off effect for a relatively smooth PV. A simple ladder
program can convert the D4-454’s continuous loop output to on/off control.

PID Loop A mathematical method of closed-loop control involving the sum of three terms based on
proportional, integral, and derivative error values. The three terms have independent gain constants,
allowing one to optimize (tune) the loop for a particular physical system.
Position Algorithm The control output is calculated so it responds to the displacement (position) of the
PV from the SP (error term)
Process A manufacturing procedure which adds value to raw materials. Process control particularly refers
to inducing chemical changes to the material in process.
Process Variable (PV) A quantitative measurement of a physical property of the material in process,
which affects final product quality and is important to monitor and control.
Proportional Gain A constant that determines the magnitude of the PID proportional term in response
to the current error.
PV Absolute Alarm A programmable alarm that compares the PV value to alarm threshold values.
PV Deviation Alarm A programmable alarm that compares the difference between the SP and PV values
to a deviation threshold value.
Ramp/Soak Profile A set of SP values called a profile, which is generated in real time upon each loop
calculation. The profile consists of a series of ramp and soak segment pairs, greatly simplifying the task of
programming the PLC to generate such SP sequences.
Rate Also called differentiator, the rate term responds to the changes in the error term.
Remote Setpoint The location where a loop reads its setpoint when it is configured as the minor loop in
a cascaded loop topology.
Reset Also called integrator, the reset term adds each sampled error to the previous, maintaining a running
total called the bias.
Reset Windup A condition created when the loop is unable to find equilibrium, and the persistent error
causes the integrator (reset) sum to grow excessively (windup). Reset windup causes an extra recovery
delay when the original loop fault is remedied.
Reverse-Acting Loop A loop in which the PV increases in response to a control output decrease. In other
words, the process has a negative gain.
Sampling time The time between PID calculations. The CPU method of process control is called a
sampling controller, because it samples the SP and PV only periodically.
Setpoint (SP) The desired value for the process variable. The setpoint (SP) is the input command to the
loop controller during closed loop operation.
Soak Deviation The soak deviation is a measure of the difference between the SP and PV during a soak
segment of the Ramp/Soak profile, when the Ramp/Soak generator is active.
Step Response The behavior of the process variable in response to a step change in the SP (in closed loop
operation), or a step change in the control output (in open loop operation)
Transfer To change from one loop operational mode to another (between Manual, Auto, or Cascade).
The word “transfer” probably refers to the transfer of control of the control output or the SP, depending
on the particular mode change.
Velocity Algorithm The control output is calculated to represent the rate of change (velocity) for the PV
to become equal to the SP.

Bibliography
Fundamentals of Process Control Theory, Second Edition Application Concepts of Process Control
Author: Paul W. Murrill Author: Paul W. Murrill
Publisher: Instrument Society of America Publisher: Instrument Society of America
ISBN 1–55617–297–4 ISBN 1–55617–080–7
PID Controllers: Theory, Design, and Tuning, 2nd Edition Author: Fundamentals of Temperature, Pressure, and Flow Measurements,
K. Astrom and T Hagglund Third edition
Publisher: Instrument Society of America Author: Robert P. Benedict
ISBN 1–55617–516–7 Publisher: John Wiley and Sons
ISBN 0–471–89383–8
Process / Industrial Instruments & Controls Handbook, Fourth pH Measurement and Control, Second Edition
Edition Author: Gregory K. McMillan
Author (Editor-in-Chief): Douglas M. Considine Publisher: Instrument Society of America
Publisher: McGraw-Hill, Inc ISBN 0-07-012445-0 ISBN 1–55617–483–7
Programmable Controllers Concepts and Applications, First Edition Fundamentals of Programmable Logic Controllers, Sensors, and
Authors: C.T. Jones and L.A. Bryan Communications
Publisher: International Programmable Controls Author: Jon Stenerson
ISBN 0-915425-00-9 Publisher: Prentice Hall ISBN 0-13-726860-2
Process Control, Third Edition Instrument Engineer’s Handbook Process Measurement and Analysis, Third Edition Instrument
Author (Editor-in-Chief): Bela G. Liptak Engineer’s Handbook
Publisher: Chilton ISBN 0–8019–8242–1 Author (Editor-in-Chief): Bela G. Liptak
Publisher: Chilton ISBN 0–8019–8197–2

CH 8

Uploaded by

Copyright:

Available Formats

CH 8

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CH 8

Uploaded by

Copyright:

Available Formats

Chapter

PID Loop Operation

D4-454 PID Control

8-2 D4-454 PLC User Manual, 1st Edition, Rev. B

PID Loop Feature Specifications

Alarm Feature Specifications

D4-454 PLC User Manual, 1st Edition, Rev. B 8-3

Introduction to PID Control

71° OFF OFF

8-4 D4-454 PLC User Manual, 1st Edition, Rev. B

D4-454 PLC User Manual, 1st Edition, Rev. B 8-5

Introducing D4-454 PID Control

8-6 D4-454 PLC User Manual, 1st Edition, Rev. B

Loop Configuring External

Setpoint Value Error Term Loop Control Output Manufacturing

D4-454 PLC User Manual, 1st Edition, Rev. B 8-7

Process Control Definitions

8-8 D4-454 PLC User Manual, 1st Edition, Rev. B

PID Loop Operation

Position Form of the PID Equation

D4-454 PLC User Manual, 1st Edition, Rev. B 8-9

Reset Windup Protection

8-10 D4-454 PLC User Manual, 1st Edition, Rev. B

Mxn = Mx if 0 < M < 1

Adjusting the Bias

Mxn = Mx if 0 < M < 1

D4-454 PLC User Manual, 1st Edition, Rev. B 8-11

Step Bias Proportional to Step Change in SP

Eliminating Proportional, Integral or Derivative Action

Velocity Form of the PID Equation

8-12 D4-454 PLC User Manual, 1st Edition, Rev. B

D4-454 PLC User Manual, 1st Edition, Rev. B 8-13

Loop Operating Modes

Special Loop Calculations

8-14 D4-454 PLC User Manual, 1st Edition, Rev. B

Error Deadband Control

D4-454 PLC User Manual, 1st Edition, Rev. B 8-15

Ten Steps to Successful Process Control

8-16 D4-454 PLC User Manual, 1st Edition, Rev. B

Step 5: Wiring and Installation

Step 10: Save Parameters

D4-454 PLC User Manual, 1st Edition, Rev. B 8-17

PID Loop Setup

PID Error Flags, V7642

PID ERROR Flags

8-18 D4-454 PLC User Manual, 1st Edition, Rev. B

Establishing the Loop Table Size and Location

V–Memory User Data

D4-454 PLC User Manual, 1st Edition, Rev. B 8-19

8-20 D4-454 PLC User Manual, 1st Edition, Rev. B

Loop Table Word Definitions

D4-454 PLC User Manual, 1st Edition, Rev. B 8-21

PID Mode Setting 1 Bit Descriptions (Addr + 00)

Bit PID Mode Setting 1 Description Read/Write Bit=0 Bit=1

8-22 D4-454 PLC User Manual, 1st Edition, Rev. B

PID Mode Setting 2 Bit Descriptions (Addr + 01)

Bit PID Mode 2 Word Description Read/Write Bit=0 Bit=1

NOTE 1: Bit 7 can be used to cancel Autotune mode by setting it to 0