C HAPTER

HAPTER
PID LOOP OPERATION
8
In This Chapter...
DL06 PID Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–2
Introduction to PID Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–4
Introducing DL06 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 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–40
Using the Special PID Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–53
Ramp/Soak Generator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–58
DirectSOFT Ramp/Soak Example . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–63
Cascade Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–65
Time-Proportioning Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–68
Feedforward Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–70
PID Example Program . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–72
Troubleshooting Tips . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–75
Glossary of PID Loop Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . .8–77
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8–79
 Chapter 8: PID Loop Operation

1 DL06 PID Control

DL06 PID Control Features
2 Along with control functions discussed in this manual, the DL06 PLC features PID process
control capability. The DL06 PID process control loops offer the same features offered in
3 much larger PLCs. The primary features are:
• Up to 8 PID loops, individual programmable sample rates
4 • Manual, Automatic and Cascade loop operation modes
• Two types of bumpless transfer available
5 • Full-featured alarms
• Ramp/soak generator with up to 16 segments
6 • Auto Tuning
The DL06 CPU has process control loop capability in addition to ladder program execution.
7 You can select and configure up to eight loops. All sensor and actuator wiring connects
directly to DL06 analog modules. All process variables, gain values, alarm levels, etc.,
associated with each loop reside in a Loop Variable Table in the CPU. The DL06 CPU reads
8 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
9 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.
10 Analog Input

11 DL06
PID Loop Calculations

12
13
14
Manufacturing Process

A
B Analog Output

DirectSOFT programming software, release 5, or later, is used for configuring analog control
C loops in the DL06. DirectSOFT 5 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 DL06’s V-memory
D (RAM). The loop parameters also may be saved to disk for recall later.

8–2 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

PID Loop Feature Specifications

Number of loops
CPU V-memory needed
Selectable, 8 maximum
32 words (V locations) per loop selected, 64 words if using ramp/soak
1
PID algorithm Position or Velocity form of the PID equation
Control Output polarity Selectable direct-acting or reverse-acting 2
Error term curves Selectable as linear, square root of error, and error squared
Loop update rate (time between
PID calculation) 0.05 to 99.99 seconds, user programmable 3
0.05 seconds for 1 to 4 loops
Minimum loop update rate
Loop modes
0.1 seconds for 5 to 8 loops
Automatic, Manual (operator control), or Cascade control
4
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) 5
Set Point Limits Specify minimum and maximum setpoint values
Process Variable Limits
Proportional Gain
Specify minimum and maximum Process Variable values
Specify gains of 0.01 to 99.99
6
Integrator (Reset) Specify reset time of 0.1 to 99.99 in units of seconds or minutes
Derivative (Rate) Specify the derivative time from 0.01 to 99.99 seconds 7
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.
8
Automatically sets the bias equal to the control output when control switches from manual to
Bumpless Transfer II
Step Bias
automatic.
Provides proportional bias adjustment for large setpoint changes
9
For position form of PID, this inhibits integrator action when the control output reaches 0% or
Anti-windup (Freeze Bias) 100 % (speeds up loop recovery when output recovers from saturation)
Specify a tolerance (plus and minus) for the error term (SP–PV), so that no change in control
10
Error Deadband output value is made
11
Alarm Feature Specifications
PV Alarm Hysteresis
PV Alarm Points
Specify 1 to 200 (word/binary) does not affect all alarms, such as PV Rate-of-Change Alarm
Select PV alarm settings for Low–low, Low, High, and High-high conditions
12
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 13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–3

 Chapter 8: PID Loop Operation

1 Introduction to PID Control

What is PID Control?
2 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.
3 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
4 throughout industry:
1. The ON-OFF controller, sometimes referred to as an open loop controller.
5 2. The PID controller, sometimes called a closed loop controller.
Regardless of type, analog controllers require input signals from electronic sensors such as
6 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.
7 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
8 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
9 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
10 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
11 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.
12
13 71°
OFF OFF

14 69° SETPOINT

A 67°
ON ON ON

B TIME

The ON-OFF controller is used in some industrial control applications, but is not practical
C in the majority of industrial control processes.
The most common process controller that is used in industry is the PID controller.
D

8–4 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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,
1
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 2
the load, thus causing a change in the process variable.
The PID controller receives signals from sensors and computes corrective action to the 3
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). 4
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
5
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 70 mph, the 6
setpoint. Now, the car is cruising at a steady 70 mph 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 70 mph without jerking
7
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 8
fuel to the engine, thus, the engine slows down allowing the car to maintain the 70 mph
speed. How does this application apply to PID control? Lets look at the function of P, I and
D terms:
9
• 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 10
manipulated variable is equal to the proportional gain multiplied by the error (the
activating signal). In mathematical terms: 11
Proportional action = proportional gain X error
Error = Setpoint (SP) - Process Variable (PV)
Applying this to the cruise control, the speed was set at 70 mph which is the Setpoint.
12
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 13
maintained at 70 mph, 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 14
the setpoint of 70 mph. 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
A
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. B
• 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
C
in advance.
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–5

 Chapter 8: PID Loop Operation

1 Introducing DL06 PID Control

The DL06 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
2 (setpoint) as long as there are no disturbances (cold water) in the process.
The DL06 PLC has the ability to directly accept signals from electronic sensors, such as
3 thermocouples, pressure, VFDs, etc. These signals may be used in mathematically derived
control systems.
4 In addition, the DL06 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
5 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, motor control, etc.)
6 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
7 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.
8 The DL06 PID control provides feedback loops using the PID algorithm. The control output
is computed from the measured process variable as follows:
Let:
9 K = proportional gain
c
T = Reset or integral time
10 i
T = Derivative time or rate
d
SP = Setpoint
11 PV(t) = Process Variable at time “t”
e(t) = SP-PV(t) = PV deviation from setpoint at time “t” or PV error.
12 Then:
13 M(t) = Control output at time “t”
t
M(t) = Kc [ e(t) + 1/T 兰 e(x) dx + T d/dt e(t) ] + M
14 i 0 d o

The analog input module receives the process variable in analog form along with an operator
A 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
B 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
C 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
D used to provide the desired control action in Proportional (P), Proportional-Integral (PI), or
Proportional-Integral-Derivative (PID) control fashion.

8–6 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Standard DL05/06 analog input modules are used to interface to field transmitters to obtain
the PV. These transmitters normally provide a 4-20mA current or an analog voltage of various
ranges for the control loop.
1
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 2
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 3
current output module to control a valve.
With DirectSOFT, additional ladder logic programming, both time proportioning (eg.,
heaters for temperature control) and position actuator (eg., reversible motor on a valve) type
4
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. 5
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. 6
7
Loop Configuring External

PLC System
and Monitoring Disturbances 8
Setpoint Value
+ k
Error Term Loop
Calculation
Control Output Manufacturing
Process
9
–
Process Variable 10
11
12
13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–7

 Chapter 8: PID Loop Operation

Process Control Definitions

1 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
2 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
3 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
4 correct, the process variable will equal the setpoint.
Control Output – The result of the loop calculation, which becomes a command for the
5 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
6 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.
7 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.
8 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.
9 Final Control Element – This is the physical device used to control the manipulated variable.
Valves are probably the most widely used final control element.
10 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
11 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
12 analyzers and quality control are usually sources of significant dead time.
Loop Configuring – Operator-initiated selections which set up and optimize the performance
13 of a control loop. The loop calculation function uses the configuration parameters in real time
to adjust gains, offsets, etc.
14 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).
A
B
C
D

8–8 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

PID Loop Operation

The Proportional–Integral–Derivative (PID) algorithm is widely used in process control. The
1
PID method of control adapts well to electronic solutions, whether implemented in analog or
digital (CPU) components. The DL06 CPU implements the PID equations digitally by 2
solving the basic equations in software. I/O modules serve only to convert electronic signals
into digital form (or vice versa).
The DL06 uses two types of PID controls: “position” and “velocity”. These terms usually refer
3
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
4
displacement (position) of the PV from the SP (error term).
• PID Velocity Algorithm – The control output is calculated to represent the rate of change 5
(velocity) for the PV to become equal to the SP.
Position Form of the PID Equation 6
Referring to the control output equation on page 8-6, the DL06 CPU approximates the
output M(t) using a discrete position form of the PID equation. 7
Let:
Ts = Sample rate 8
Kc = Proportional gain
Ki = Kc * (Ts/Ti) = Coefficient of integral term
Kr = Kc * (Td/Ts) = Coefficient of derivative term 9
Ti = Reset or integral time
Td = Derivative time or rate
SP = Setpoint
10
PVn = Process variable at nth sample
en = SP – PVn = Error at nth sample
Mo = Value to which the controller output has been initialized
11
Then:
Mn = Control output at nth sample
12
n
Mn = Kc * en + Ki 兺 ei + Kr (en - en-1) + Mo
13
i=1

This form of the PID equation is referred to as the position form since the actual actuator
14
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 A
the process variable instead of the error as follows:
n
B
Mn = Kc * en + Ki 兺 ei + Kr (PVn - PVn-1) + Mo
i=1
C
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 D
resulting in a bump to the process due to derivative action. This bump is not present in the
second form of the equation.

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–9

 Chapter 8: PID Loop Operation

The DL06 also combines the integral sum and the initial output into a single term called the
1 bias (Mx). This results in the following set of equations:
Mxo = Mo
2 Mx = Ki * en + Mxn-1
Mn = Kc * en - Kr(PVn-PVn-1) + Mxn
3 The DL06 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 DL06 also allows you to specify the minimum and maximum output limit
4 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
5 these equations can be found in most good text books about process control.

6 Reset Windup Protection

Reset windup can occur if reset action (integral term) is specified and the computation of the
7 bias term Mx is:
Mx = Ki * en + Mxn-1
8 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
9 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
10 for long enough to cause the bias to become positive again. This will cause the process
variable to undershoot.
11 One way to solve the problem is to simply clamp the normalized bias between 0.0 and 1.0.
The DL06 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
12 also cause the process variable to undershoot.
The DL06 CPU is programmed to solve the overshoot problem by either freezing the bias
13 term, or by adjusting the bias term.

14
A
B
C
D

8–10 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Freeze Bias
If the “Freeze Bias” option is selected when setting up the PID loop (discussed later) then the 1
CPU simply stops changing the bias (Mx) whenever the computed normalized output (M)
goes outside the interval 0.0 to 1.0.
Mx = Ki * en + Mxn-1
2
M = Kc * en - Kr(PVn - PVn-1) + Mx
3
Mn = 0 if M < 0
4
Mn = M if 0 ⭐ M ⭐ 1
Mn = 1 if M > 1 5
Mxn = Mx if 0 ⭐ M ⭐ 1 6
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
7
the bias had been allowed to go all the way to zero. This action has the effect of reducing the
amount of overshoot. 8
Adjusting the Bias 9
The normal action of the CPU is to adjust the bias term when the output goes out of range as
shown below.
Mx = Ki * en + Mxn-1 10
M = Kc * en - Kr(PVn - PVn-1) + Mx
11
Mn = 0 if M < 0
Mn = M if 0 ⭐ M ⭐ 1 12
Mn = 1 if M > 1
13
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
14
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 A
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 B
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). C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–11

 Chapter 8: PID Loop Operation

Step Bias Proportional to Step Change in SP

1 This feature reduces oscillation caused by a step change in setpoint when the adjusting bias
feature is used.
2 Mx = Mx * SPn / SPn-1 if the loop is direct acting
Mx = Mx * SPn-1 / SPn if the loop is reverse acting
3
Mxn = 0 if Mx < 0
4 Mxn = Mx if 0 ⭐ Mx ⭐ 1
Mxn = 1 if M > 1

5 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
6 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.
7 Eliminating Integral Action The effect of integral action on the output may be
eliminated by setting Ti = 9999. When this is done, the
8 user may then manually control the bias term (Mx) to
eliminate any steady-state offset.
9 Eliminating Derivative Action The effect of derivative action on the output may be
eliminated by setting Td = 0 (most loops do not require a
D parameter; it may make the loop unstable).
10 Eliminating Proportional Action Although rarely done, the effect of proportional term on
the output may be eliminated by setting Kc = 0. Since Kc
11 is also normally a multiplier of the integral coefficient (Ki)
and the derivative coefficient (Kr), the CPU makes the
12 computation of these values conditional on the value of
Kc as follows:
Ki = Kc * (Ts / Ti) if Kc ⫽ 0
13 Ki = Ts / Ti if Kc = 0 (I or ID only)
14 Kr = Kc * (Td / Ts) if Kc ⫽ 0
Kr = Td / Ts if Kc = 0 (ID or D only)
A
Velocity Form of the PID Equation
B 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
C 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”.
D The velocity equation is given by:
⌬Mn = M - Mn-1
⌬Mn = Kc * (en - en-1) + Ki * (PVn - 2 * PVn-1 + PVn-2)

8–12 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Bumpless Transfer
The DL06 loop controller provides for bumpless mode changes. A bumpless transfer from 1
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
2
initialized as follows:
Position PID Algorithm Velocity PID Algorithm
3
SP = PV SP = PV
Mx = M
4
The bumpless transfer feature of the DL06 is available in two types: Bumpless I and Bumpless
II (see page 8-26). The transfer type is selected when the loop is set up.
5
Loop Alarms 6
The DL06 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: 7
• PV Limit – Specify up to four PV alarm points.
High-High PV rises above the programmed High-High Alarm Limit. 8
High PV rises above the programmed High Alarm Limit.
Low
Low-Low
PV fails below the Low Alarm Limit.
PV fails below the Low-Low Limit.
9
• 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 10
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.
11
• Rate of Change – This alarm is set when the PV changes faster than a specified rate-of-
change limit. 12
• 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
13
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 14
process operators. The solution is to use the PV Alarm Hysteresis feature.
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–13

 Chapter 8: PID Loop Operation

Loop Operating Modes

1 The DL06 loop controller operates in one of three modes, either Manual , Automatic or
Cascade.
2 Manual
In manual mode, the control output is determined by the operator, not the loop controller.
3 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.
4 Automatic
In automatic mode, the loop controller computes the control output based on the
5 programmed parameters stored in V-memory. All alarms are monitored while in automatic.
Cascade
6 Cascade mode is an option with the DL06 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.
7 Special Loop Calculations
8 Reverse Acting Loop
Although the PID algorithm is used in a direct, or forward, acting loop controller, there are
9 times when a reverse acting control output is needed. The DL06 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
10 the output to increase the PV.
Mx = -Ki * en + Mxn-1
11 M = -Kc * en + Kr(PVn-PVn-1) + Mxn
Square Root of the Process Variable
12 Square root is selected whenever the PV is from a device such as an orifice meter which
requires this calculation.
13 Error Squared Control
Whenever error squared control is selected, the error is calculated as:
14 en = (SP - PVn) * ABS(SP - PVn)
A loop using the error squared is less responsive than a loop using just the error, however, it
A 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.
B
C
D

8–14 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Error Deadband Control

With error deadband control, no control action is taken if the PV is within the specified 1
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
2
the loop calculation.
en = 0 SP - Deadband_Below_SP < PV < SP - Deadband_Above_SP
3
en = P - PVn otherwise
The error will be squared first if both Error Squared and Error Deadband is selected. 4
Derivative Gain Limiting
When the coefficient of the derivative term, Kr, is a large value, noise introduced into the PV
5
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 6
term computation, Yn, as shown below.

Y n = Yn-1 +
Ts 7
* (PV n - Y n-1 )
Ts + ( Td )
Kd 8
Position Algorithm 9
Mx = Ki * en + Mxn-1
M = Kc * en - Kr * (Yn-Yn-1) + Mx 10
Velocity Algorithm
⌬M = Kc * (en - en-1) + Ki * en - Kr * (Yn - 2 * Yn-1 + Yn-2) 11
12
13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–15

 Chapter 8: PID Loop Operation

1 Ten Steps to Successful Process Control

Controllers such as the DL06 PLC provide sophisticated process control features. Automated
control systems can be difficult to debug, because a given symptom can have many possible
2 causes. We recommend a careful, step-by-step approach to bringing new control loops online:
Step 1: Know the Recipe
3 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:
4 • Identify all relevant Process Variables, such as temperature, pressure, or flow rates, etc.,
which need precise control.
5 • Plot the desired Setpoint values for each process variables for the duration of one process
cycle.
6 Step 2: Plan Loop Control Strategy
This simply means choosing the method the machine will use to maintain control over the
7 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
8 whether a machine operator can change the SP.
Step 3: Size and Scale Loop Components
9 Assuming the control strategy is sound, it is still crucial to properly size the actuator and
properly scale the sensors.
10 • 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
11 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.
12 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
13 level). However, an over-sensitive sensor can cause control oscillations, etc. The DL06
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.
14 Step 4: Select I/O Modules
After deciding the number of loops, PV variables to measure, and SP values, you can choose
A the appropriate I/O module. Refer to the figure on the next page. In many cases, you will be
able to share input or output modules, or use an analog I/O combination module, among
B several control loops. The example shown sends the PV and Control Output signals for two
loops through the same set of modules.

C Automationdirect offers DL06 analog input modules with 4 channels per module that accept
0 – 20mA or 4 – 20mA signals. Also, analog input and output combination modules are now
available. Thermocouple and RTD modules can also be used to maintain temperatures to a
D 10th of a degree. Refer to the sales catalog for further information on these modules, or find
the modules on our website, www.automationdirect.com.

8–16 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

DL06 CPU
Input V-memory Digital
Module Output

Channel 1 PV
Loop 1 Data
SP
Loop 2 Data
OUT Channel 1 Process 1 1
Channel 2 PV SP OUT Channel 2 Process 2

Channel 3
2
Channel 4
3
Step 5: Wiring and Installation
After selection and procurement of all loop components and I/O module(s), you can
4
perform the wiring and installation. Refer to the wiring guidelines in Chapter 2 of this
manual, and to the D0–OPTIONS–M manual. The most common wiring errors when
installing PID loop controls are:
5
• Reversing the polarity of sensor or actuator wiring connections.
• Incorrect signal ground connections between loop components.
6
Step 6: Loop Parameters 7
After wiring and installation, choose the loop setup parameters. The easiest method for
programming the loop tables is using DirectSOFT (5.0 or later). This software provides PID
Setup using dialog boxes to simplify the task. Note: It is important to understand the
8
meaning of all loop parameters mentioned in this chapter before choosing values to enter.
Step 7: Check Open Loop Performance
9
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. 10
• 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 11
responds (and moves in the correct direction!).
Step 8: Loop Tuning 12
If the Open Loop Test (page 8–40) 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 13
page 8–45). In this step, the loop is tuned so the PV automatically follows the SP.
Step 9: Run Process Cycle 14
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
A
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
B
the process goes out of control. Damage to equipment and/or serious injury to personnel can
result from loss of control of some processes. C
Step 10: Save Parameters D
When the loop tests and tuning sessions are complete, be sure to save all loop setup
parameters to disk.

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–17

 Chapter 8: PID Loop Operation

PID Loop Setup

1 Some Things to Do and Know Before Starting
Have your analog module installed and operational before beginning the loop setup (refer to
2 the DL05/06 Option Modules User Manual, D0-OPTIONS-M). The DL06 PLC gets its
PID loop processing instructions from V-memory tables. There isn’t a PID instruction that
3 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.
4 Address Setup Parameter Data type Ranges Read/Write
Loop Parameter V1200 – V7340
5 V7640
V7641
Table Pointer
Number of Loops
Octal
BCD
V10000-V17740
1–8
write
write
V7642 Loop Error Flags BITS 0 or 1 read
6
NOTE: The V-memory data is stored in SRAM memory. If power is removed from the CPU for an extended
7 period of time, the PID Setup Parameters will be lost. It is recommended to use the optional battery backup
to retain the memory in SRAM. Another option is to use the MOV instruction, which places the data in
non-volatile memory, when setting up the parameters in the ladder program.
8 PID Error Flags, V7642
PID Error Flags
9 The CPU reports any programming errors of the setup parameters in V7640 and V7641. It
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

does this by setting the appropriate bits in V7642 on program-to-run mode transitions.
10 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
11 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
12 errors.
Bit Error Description (0 = no error, 1 = error)
13 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.
14 2
3
The number of loops selected (in V7641) is greater than 8.
The loop table extends past (straddles) the boundary at V7377. Use an address closer to V1200.

A 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
B 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
C parameters the CPU will read or write on every loop calculation.
CPU Tasks V–Memory Space

D Ladder
Program
READ/
WRITE
User Data

LOOP
DATA
CONFIGURE/
MONITOR
PID Loop
Task
Setup Parameters
READ V7640, V7641
(at powerup)
DirectSOFT32 Programming Software

8–18 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

NOTE: The DL06 CPU’s PID algorithm requires DirectSOFT Version 5.0 (or later) and firmware
version 2.1 (or later). See our website for more information: www.automationdirect.com. 1
The Loop Table contains data for only the number of loops
that are selected. The address for the table is stored in V7641.
V–Memory User Data
2
Each loop configuration occupies 32 words (0 to 37 octal) in
the loop table.
V2000
V2037
LOOP #1
32 words
3
For example, consider an application with 4 loops, and V2000 V2040 LOOP #2
has been chosen as the starting location. The Loop Parameter
will occupy V2000 – V2037 for loop 1, V2040 – V2077 for
V2077 32 words
LOOP #3
4
32 words
loop 2 and so on. Loop 4 occupies V2140 - V2177.
Determine the block of V-memory to be used for each PID
LOOP #4
32 words
5
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 6
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.
7
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.
8
Using DirectSOFT is the simplest way to setup the parameters. To setup the PID parameters, 9
the DL06 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
10
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 11
in the diagram below. After the address has been entered, the memory range will appear. Also,
entering the number of PID loops (1 to 8) 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
12
for a basic loop operation for each loop can be setup with the dialogs made available.
13
14
A
B
C
NOTE: Have an edited program open, then click on PLC > Setup > PID to access the Setup PID
dialog.
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–19

 Chapter 8: PID Loop Operation

Loop Table Word Definitions

1 These are the loop parameters associated with each of the four loops available in the DL06.
The parameters are listed in the following table. The address offset is in octal, to help you
2 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.
3
Read on-
Word # Address+Offset Description Format the- fly***
4 1 Addr + 0 PID Loop Mode Setting 1 bits Yes
2 Addr + 1 PID Loop Mode Setting 2 bits Yes
5 3
4
Addr + 2
Addr + 3
Setpoint Value (SP)
Process Variable (PV)
word/binary
word/binary
Yes
Yes

6 5
6
Addr + 4
Addr + 5
Bias (Integrator) Value
Control Output Value
word/binary
word/binary
Yes
Yes
7 Addr + 6 Loop Mode and Alarm Status bits –
7 8
9
Addr + 7
Addr + 10
Sample Rate Setting
Gain (Proportional) Setting
word/BCD
word/BCD
Yes
Yes

8 10
11
Addr + 11
Addr + 12
Reset (Integral) Time Setting
Rate (Derivative) Time Setting
word/BCD
word/BCD
Yes
Yes
12 Addr + 13 PV Value, Low-low Alarm word/binary No*
9 13
14
Addr + 14
Addr + 15
PV Value, Low Alarm
PV Value, High Alarm
word/binary
word/binary
No*
No*

10 15
16
Addr + 16
Addr + 17
PV Value, High-high Alarm
PV Value, deviation alarm (YELLOW)
word/binary
word/binary
No*
No*
17 Addr + 20 PV Value, deviation alarm (RED) word/binary No*
11 18
19
Addr + 21
Addr + 22
PV Value, rate-of-change alarm
PV Value, alarm hysteresis setting
word/binary
word/binary
No*
No*

12 20
21
Addr + 23
Addr + 24
PV Value, error deadband setting
PV low-pass filter constant
word/binary
word/BCD
Yes
Yes
22 Addr + 25 Loop derivative gain limiting factor setting word/BCD No**
13 23
24
Addr + 26
Addr + 27
SP value lower limit setting
SP value upper limit setting
word/binary
word/binary
Yes
Yes

14 25
26
Addr + 30
Addr + 31
Control output value lower limit setting
Control output value upper limit setting
word/binary
word/binary
No**
No**
27 Addr + 32 Remote SP Value V-Memory Address Pointer word/hex Yes
A 28
29
Addr + 33
Addr + 34
Ramp/Soak Setting Flag
Ramp/Soak Programming Table Starting Address
bit
word/hex
Yes
No**

B 30
31
Addr + 35
Addr + 36
Ramp/Soak Programming Table Error Flags
PV auto transfer, channel number
bits
word/hex
No**
Yes
32 Addr + 37 Control output auto transfer, channel number word/hex Yes
C * Read data only when alarm enable bit transitions from 0 to 1.
** Read data only on PLC Mode change.
D *** Read on-the-fly means that the content of V-memory can be changed while the PID loop is in
operation.

8–20 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop 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 1
following table.
Bit PID Mode Setting 1 Description Read/Write Bit=0 Bit=1 2
0 Manual Mode Loop Operation request write – 01 request
1
2
Automatic Mode Loop Operation request
Cascade Mode Loop Operation request
write
write
–
–
01 request
01 request
3
3 Bumpless Transfer select write Mode I Mode II
4 Direct or Reverse-Acting Loop select write Direct Reverse 4
5 Position / Velocity Algorithm select write Position Velocity
6
7
PV Linear / Square Root Extract select
Error Term Linear / Squared select
write
write
Linear
Linear
Sq. root
Squared
5
8 Error Deadband enable write Disable Enable
9 Derivative Gain Limit select write Off On 6
10 Bias (Integrator) Freeze select write Off On
11
12
Ramp/Soak Operation select
PV Alarm Monitor select
write
write
Off
Off
On
On
7
13 PV Deviation alarm select write Off On
14 PV rate-of-change alarm select write Off On 8
15 Loop mode is independent from CPU mode when set write Loop with CPU Loop Independent
mode of CPU mode
9
10
11
12
13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–21

 Chapter 8: PID Loop Operation

PID Mode Setting 2 Bit Descriptions (Addr + 01)

1 The individual bit definitions of the PID Mode Setting 2 word (Addr+01) are listed in the
following table.
2 Bit PID Mode 2 Word Description Read/Write Bit=0 Bit=1
0 Input (PV) and Control Output Range Unipolar/Bipolar write unipolar bipolar
3 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 on
4 3 SP Input limit enable write disable enable
4 Integral Gain (Reset) units select write seconds minutes

5 5 Select Auto tune PID algorithm

6 Auto tune selection
write
write
closed loop
PID
open loop
PI only (rate = 0)
7 Auto tune start read/write auto tune force start
6 (See Note 1)
8 PID Scan Clock (internal use) read
cancel/done
– –
9 Input/Output Data Format 16-bit select
7 (See Notes 1 and 2) write not 16 bit select 16 bit

10 Select separate data format for input and output

(See Notes 2, and 3) write same format separate formats
8 11 Control Output Range Unipolar/Bipolar select
See Notes 2, and 3) write unipolar bipolar
12 Output Data Format select (See Notes 2, and 3) write 12 bit 15 bit
9 13 Output data format 16-bit select (See Notes 2, and 3) write not 16 bit select16 bit
14–15 Reserved for future use – – –
10
NOTE 1: Bit 7 can be used to cancel Autotune mode by setting it to 0.
11
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,
12 then the values in bits 0 and 1 are not read, and bit 9 defines the data format (the range is
automatically unipolar).

13 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,
14 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.
A 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
B not read, and bit 13 defines the data format, (the output range is automatically unipolar).

C
D

8–22 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Mode/Alarm Monitoring Word (Addr + 06)

The individual bit definitions of the Mode / Alarm monitoring (Addr+06) word are listed in 1
the following table.
Bit Mode/Alarm Bit Description
0 Manual Mode Indication
Read/Write
read
Bit=0
–
Bit=1
Manual
2
1 Automatic Mode Indication
2 Cascade Mode Indication
read
read
–
–
Auto
Cascade 3
3 PV Input LOW–LOW Alarm read Off On
4 PV Input LOW Alarm
5 PV Input HIGH Alarm
read
read
Off
Off
On
On
4
6 PV Input HIGH–HIGH Alarm
7 PV Input YELLOW Deviation Alarm
read
read
Off
Off
On
On 5
8 PV Input RED Deviation Alarm read Off On
9 PV Input Rate-of-Change Alarm
10 Alarm Value Programming Error
read
read
Off
–
On
Error
6
11 Loop Calculation Overflow/Underflow
12 Loop in Auto-Tune indication
read
read
–
Off
Error
On 7
13 Auto-Tune error indication read – Error
14–15 Reserved for Future Use – – – 8
Ramp/Soak Table Flags (Addr + 33) 9
The individual bit definitions of the Ramp/Soak Table Flag (Addr+33) word are listed in the
following table.
10
Bit Ramp/Soak Flag Bit Description Read/Write Bit=0 Bit=1
0
1
Start Ramp / Soak Profile
Hold Ramp / Soak Profile
write
write
–
–
01 Start
01 Hold
11
2 Resume Ramp / soak Profile write – 01 Resume
3 Jog Ramp / Soak Profile write – 01 Jog 12
4 Ramp / Soak Profile Complete read – Complete
5
6
PV Input Ramp / Soak Deviation
Ramp / Soak Profile in Hold
read
read
Off
Off
On
On
13
7 Reserved read – –
8–15 Current Step in R/S Profile read decode as byte (hex) 14
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,
A
F, and 10, which represent segments 1 to 16 respectively. If the byte=0, then the Ramp/Soak
table is not active. B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–23

 Chapter 8: PID Loop Operation

Ramp/Soak Table Location (Addr + 34)

1 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
2 Soak feature, you must program a separate table of 32 words with appropriate values. The
DirectSOFT dialog box makes this easy to do.

3 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,
4 regardless of the number of segments programmed.
The ramp/soak table parameters are defined in the table below. Further details are in the
5 section on Ramp/Soak Operation in this chapter.

6 Addr
Offset Step Description Addr
Offset Step Description
+ 00 1 Ramp End SP Value + 20 9 Ramp End SP Value
7 V–Memory Space

User Data
+ 01 1 Ramp Slope + 21 9 Ramp Slope
+ 02 2 Soak Duration + 22 10 Soak Duration
8 V2000
V2037
LOOP #1
32 words
+ 03
+ 04
2
3
Soak PV Deviation
Ramp End SP Value
+ 23
+ 24
10
11
Soak PV Deviation
Ramp End SP Value
LOOP #2
32 words + 05 3 Ramp Slope + 25 11 Ramp Slope
9 + 06 4 Soak Duration + 26 12 Soak Duration
V3000 Ramp/Soak #1
+ 07 4 Soak PV Deviation + 27 12 Soak PV Deviation
10 32 words + 10
+ 11
5
5
Ramp End SP Value
Ramp Slope
+ 30
+ 31
13
13
Ramp End SP Value
Ramp Slope
+ 12 6 Soak Duration + 32 14 Soak Duration
11 V2034 = 3000 Octal + 13 6 Soak PV Deviation + 33 14 Soak PV Deviation
Pointer to R/S table + 14 7 Ramp End SP Value + 34 15 Ramp End SP Value
12 + 15
+ 16
7
8
Ramp Slope
Soak Duration
+ 35
+ 36
15
16
Ramp Slope
Soak Duration
+ 17 8 Soak PV Deviation + 37 16 Soak PV Deviation
13
Ramp/Soak Table Programming Error Flags (Addr + 35)
14 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
A section and in the PV Alarm section later in chapter 8.

Bit R/S Error Flag Bit Description Read/Write Bit=0 Bit=1

B 0 Starting Addr out of lower V-memory range read – Error
1 Starting Addr out of upper V-memory range read – Error
C 2–3
4
Reserved for Future Use
Starting Addr in System Parameter V-memory Range
–
read
–
–
–
Error
5–15 Reserved for Future Use – – –
D

8–24 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Configure the PID Loop

Once the PID table is established in V-memory, configuring the PID loop continues with the 1
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.
2
3
4
5
6
7
Select the Algorithm Type 8
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 9
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
10
The main tasks of the CPU fall into categories as shown to the right.
The list represents the tasks done when the CPU is in Run Mode, on Read
11
Inputs
each PLC scan. Note that PID loop calculations occur after the ladder
logic task. Service
Peripherals
12
The sample rate of a control loop is simply the frequency of the PID
calculation. Each calculation generates a new control output value. With Ladder
13
the DL06 CPU, you can set the sample rate of a loop from 50 ms to PLC
Program

99.99 seconds. Most loops do not require a fresh PID calculation on

every PLC scan. Some loops may need to be calculated only once in
Scan
Calculate
PID Loops
14
1000 scans.
Enter 0.05 sec., or the sample rate of your choice, for each loop, and the Internal
Diagnostics
A
CPU automatically schedules and executes PID calculations on the
appropriate scans. Write
Outputs
B
NOTE: If more than 4 loops are programmed, enter a minimum of 0.1 second.
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–25

 Chapter 8: PID Loop Operation

Select Forward/Reverse
It is important to know which direction the control output will respond to the error (SP-PV),
1 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
2 PID loops are forward acting, such as a heating control loop. An increase in heat applied will
increase the PV (temperature).
3 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).
4 The Transfer Mode
Choose either Bumpless I or Bumpless II to provide a smooth transition of the control output
5 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
6 BumplessII.
The characteristics of Bumpless I and II transfer types are listed in the chart below. Note that
7 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.
8 Transfer Manual-to-Auto Auto-to-Cascade
Transfer Type Select PID Algorithm
Bit 3 Transfer Action Transfer Action
9 Bumpless
Position Forces Bias = Control Output Forces Major Loop Output =
Forces SP = PV Minor Loop PV
Transfer I 0
Forces Major Loop Output =
10 Velocity
Position
Forces SP = PV
Forces Bias = Control Output
Minor Loop PV
none
Bumpless 1
Transfer II
11 Velocity none none

The transfer type can also be selected in an RLL program by setting bit 3 of PID Mode 1,
12 V+00 setting as shown.

Bumpless Transfer I/II Select

PID Mode 1 Setting V+00
13 Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

SP/PV & Output Format

14 Bumpless Transfer I / II select
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
A 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.
B The reason for this is that they become independently selectable in the SP/PV and the Output
dialogs.
C 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
D 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.

8–26 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Setpoint V+02 Loop Control Output V+05

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

Select data
00 12 bit unipolar 0 to 0FFF (0 to 4095)

0 to 0FFF, 8FFF to 8001

4
format using 01 12 bit bipolar
(0 to 4095, *–4095 to 4095)
bits 0 and 1.
10 15 bit unipolar 0 to 32767 5
11 15 bit bipolar 0 to 7FFF, FFFF to 8001
= sign bit
(0 to 32767, *–32767 to 32767) 6
* Magnitude plus sign

The data format determines the numerical interface between the PID loop and the PV sensor,
7
and the control output device. This selects the data format for both the SP and the PV.
Loop Mode 8
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 9
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. 10
The DL06 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. 11
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
12
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 13
schematic diagram of manual mode operation.
Input from Operator Manual 14
Control Output V+05
Loop
Calculation
A
Auto
In Automatic Mode, the loop operates normally and generates new control output values. It B
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. C
Input from Operator Manual
Control Output V+05 D
Loop
Calculation
Auto

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–27

 Chapter 8: PID Loop Operation

In Cascade Mode, the loop operates as it does in Automatic Mode, with one important
1 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
2 output from another loop’s parameter table, V+05.
Another loop Cascaded loop
3 Control Output V+05 Cascade
Loop
Calculation Control Output
4 Normal SP V+02
Setpoint
+ k
Loop
Calculation
–
Auto/Manual
5 Process Variable

As pictured below, a loop can be changed from one mode to another, but cannot go from
6 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.
7 Manual Automatic Cascade

8 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.
9 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
10 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.
11 CPU Modes:

12 Program
Mode change
Run

13 0 = loop follows PLC mode

Loop Mode Linking
PID Mode 1 Setting V+00
1 = loop is independent
14 of PLC mode
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

Loop
Modes:
A Manual
Mode change
Automatic
Mode change
Cascade

B
If bit 15 is set to one, then the loops will run independently of the CPU mode. It is like
C having two independent processors in the CPU... one is running the RLL program and the
other is running the process loops.

8–28 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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
1
the feature used for this.
If you need to operate the PID loops while the RLL program is halted, in Program Mode,
2
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 3
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. 4
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
5
Variable. There is a block within this block called
Auto Transfer From (from analog input) with the 6
information grayed out. Checking the box to the left
of the Auto Transfer From will highlight the 7
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.
8
The second choice is V-Memory. When this is
selected, the V-memory address from where the PV 9
is transferred must be specified.
Whichever method of auto transfer is used, it is 10
recommended to check the Enable Filter Factor (a
low pass filter) and specify the coefficient. 11
You should also select the analog output for the control output to 12
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
13
checked, enter the slot number where the output module is
residing and then enter the analog output channel number. 14
A
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, B
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,
C
restore bit 15 to a value of 1 to re-establish PID operation independent of CPU.
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–29

 Chapter 8: PID Loop Operation

You may optionally configure each loop to access its analog I/O (PV and control output) by
1 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.
2
3 Setpoint V +02
k
Error Loop
Calculation
Control output V+05

4 Process variable V+03

5 V2036
Loop Table
0X XX Base/Slot /Channel number for PV
V2037 0X XX Base/Slot/Channel number for Output

6 XX 0X
Channel number 1 to 4
7 Slot number 1 to 4
Base 0

8 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
9 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.
10 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.
11
SP/PV Addresses
12 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
13 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
14 values to be entered. Set the limits around the SP value to prevent an operator from entering a

A
B
C
D

8–30 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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
1
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. 2
NOTE: The SP/PV dialog can be left as it first appears for basic PID operation.
3
Set Control Output Limits
Another dialog that will be available in the PID setup will be the Output dialog. The control 4
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
5
PID operation using a 12-bit output module, set the Upper Limit to 4095 and leave the
6
7
8
9
10
Lower Limit set to 0. Check the box next for Auto transfer to I/O module if there is a need to
11
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 12
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
13
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). 14
WARNING: If the Upper Limit is set to zero, the output will never get above zero. In effect, there
A
will be no control output.
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–31

 Chapter 8: PID Loop Operation

Enter PID Parameters

1 Another PID setup dialog, Tuning, is for entering the PID parameters shown as: Gain
(Proportional Gain), Reset (Integral Gain) and Rate (Derivative Gain).
2
3
4
5
6
7 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
8 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
9 term receives the same error signal value.
Loop Calculation
10 Setpoint Error T erm
P
+ Control Output
+ k I + k
11 –
Process Variable
D +

12 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
13 in minutes, and Derivative gain is in seconds.
Gain (Proportional Gain) – This is the most basic gain of the three. Values range from 0000
14 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
A integral-only loops.
Reset (Integral Gain) – Values range from 0001 to 9998, but they are used internally as
B xx.xx. An entry of “0000” or “9999” causes the integral gain to be⬁⬁, 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
C shown in the above dialog.
Rate (Derivative Gain) – Values which can be entered range from 0001 to 9999, but they are
D 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.

8–32 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

NOTE: You may elect to leave the tuning dialog blank and enter the tuning parameters in the
DirectSOFT PID View. 1
Loop Calculation
P
Proportional
Control
2
Setpoint Error T erm + Output
k Integral
k
+
–
I

D
Derivative
0
+
+ 3
Process Variable
Derivative,
gain-limited
1
4
Loop Table
5
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
6
Derivative gain limit select

Derivative Gain Limiting 7

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 8
function of the gain-limiting is shown in the diagram below.
The gain limit can be particularly useful during loop tuning. Most loops can tolerate only a
little derivative gain without going into uncontrolled oscillations.
9
If this option is checked, a Limit from 0 to 20 must also be entered. 10
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
11
required to be entered in the Tuning dialog, but they can set later in the DirectSOFT PID View.
12
Error Term Selection
The error term is internal to the CPUs PID loop controller, and is generated again in each 13
PID calculation. Although its data is not directly accessible, you can easily calculate it
by subtracting: Error = (SP–PV). If the PV square-root extract is enabled, then: Error = 公PV
In any case, the size of the error and algebraic sign determine the next change of the
14
control output for each PID calculation.
Error Squared – When selected, the squared error function simply squares the error term (but
A
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 B
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
C
larger errors.
• Non-linear process – some processes (such as chemical pH control) require non-linear controllers D
for best results. Another application is surge tank control, where the Control Output signal must
be smooth.

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–33

 Chapter 8: PID Loop Operation

Enable Deadband – When selected, the enable deadband function takes a range of small error
1 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.
2 Freeze Bias
The term reset windup refers to an undesirable characteristic of integrator behavior which
3 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
4 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
5 consequently has an extended recovery time. Until recovery, the output level is wrong and
causes further problems.

6 PV
PV loss PV loss
0
7 Reset windup Freeze bias enabled
Bias
8
9 Output

Recovery time Recovery time

10 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
11 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
12 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.
13 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
14 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.

A
B
C
D

8–34 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Setup the PID Alarms

Although the setup of the PID alarms is optional, you surely would not want to operate a 1
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
2
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 3
condition has been repaired.
The alarm thresholds are fully programmable, and each type of alarm may be independently 4
enabled and monitored. The following diagram shows the Alarm dialog in the PID setup
which simplifies the alarm setup.
5
6
7
8
9
Monitor Limit Alarms 10
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 11
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,
12
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 13
threshold values for these alarms should be set to give an operator an early warning if the
process loses control. 14
High–high Alarm
High Alarm V+16 XXXX
Loop Table
High-high Alarm
A
V+15 XXXX High Alarm
PV
Low Alarm
V+14 XXXX Low Alarm
B
V+13 XXXX Low-low Alarm
Low–low Alarm
C
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.
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–35

 Chapter 8: PID Loop Operation

If the process remains out of control for some time, the PV will eventually cross one of the
1 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.
2 The PV Absolute Value Alarms are reported in the four bits in PID Mode and Alarm Status V+06
the PID Mode and Alarm Status word in the loop table, as
3 shown to the right. We highly recommend using ladder logic Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
to monitor these bits. The bit-of-word instructions make this High-high Alarm
4 easy to do. Additionally, you can monitor PID alarms using
DirectSOFT.
High Alarm
Low Alarm
Low-low Alarm

5 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
6 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
7 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.
8 Red Deviation Alarm
Yellow Deviation Alarm
Red
Yellow Loop Table

9 SP Green V+17
V+20
XXXX
XXXX
Yellow Deviation Alarm
Red Deviation Alarm
Yellow Deviation Alarm Yellow

10 Red Deviation Alarm

Red

11 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.
12 The PV Deviation Alarms are reported in the two bits in the
PID Mode and Alarm Status V+06
PID Mode and Alarm Status word in the loop table, as
13 shown to the right. We highly recommend using ladder logic Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

to monitor these bits. The bit-of-word instructions make this Red Deviation
easy to do. Additionally, you can monitor PID alarms using Yellow Deviation
14 DirectSOFT.
The PV Deviation Alarm can be independently enabled and disabled from the other PV
A 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
B absolute value alarms, and is discussed at the end of this section.

C
D

8–36 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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 1
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
2
quickly and effectively, the PV absolute value will not reach the point where the material in
process would be ruined. 3
The DL06 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 4
programmed into the loop table location V+21.

Loop Table
5
PV slope OK PV slope excessive

PV
V+21 XXXX PV Rate-of-Change Alarm
6
rate-of-change alarm
PID Mode and Alarm Status V+06
7
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
8
As an example, suppose the PV is the temperature for your process, and you want an alarm 9
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
10
to convert our engineering units to counts/sample period:
15 degrees 10 counts / degree 150
11
Alarm Rate-of-Change = X = = 5 counts / sample period
1 minute 30 loop samples / min. 30
12
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.
13
The alarm hysteresis feature (discussed next) does not affect the Rate-of-Change Alarm.
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–37

 Chapter 8: PID Loop Operation

PV Alarm Hysteresis
1 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
2 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
3 Alarm Hysteresis feature.
The PV Alarm Hysteresis amount is programmable from 1 to 200 (binary/decimal). When
4 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
5 the PV value goes past a threshold and descends back through it.

6 Alarm threshold
Hysteresis

7 PV
V+22 XXXX
Loop Table
PV Alarm Hysteresis

Alarm 1
8 0

9 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
10 turning off until the PV value has returned through the threshold by the hysteresis amount.
Alarm Programming Error
11 The PV Alarm threshold values must have
certain mathematical relationships to be valid.
PID Mode and Alarm Status V+06
The requirements are listed below. If not met,
12 the Alarm Programming Error bit will be set, as Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
indicated to the right.
13 • PV Absolute Alarm value requirements:
Alarm Programming Error

Low-low < Low < High < High-high

14 • PV Deviation Alarm requirements:
Yellow < Red

A Loop Calculation Overflow/Underflow Error

This error occurs whenever the output reaches its PID Mode and Alarm Status V+06
upper or lower limit and the PV does not reach
B the setpoint. A typical example might be when a Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
valve is stuck, the output is at its limit, but the PV
C has not reached setpoint. Loop Calculation Overflow/Underflow Error

NOTE: Overflow/underflow can be alarmed in PID View. The optional C-more operator interface panel
D (see the automationdirect.com website) can also be setup to read these error bits using the PID
Faceplate templates.

8–38 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Ramp/Soak
R/S (Ramp/Soak) is the last dialog available in the PID setup. The basic PID does not require 1
any entries to be made in order to operate the PID loop. Ramp/Soak will be discussed in
another section.
2
3
4
5
6
7
Complete the PID Setup 8
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 9
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
10
and description for the loop. This is useful if there is more than one PID loop in your
application. 11
12
13
Save to disk
14
Save to PLC A
B
NOTE: It is good practice to save your project after setting up the PID loop by selecting File from the C
menu toolbar, then Save project > to disk. In addition to saving your entire project, all the PID
parameters are also saved. D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–39

 Chapter 8: PID Loop Operation

1 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.
2 The quality of a loop’s performance may generally be judged by how well the PV follows the
SP after an SP step change. It is important to keep in mind that understanding the process is
3 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.
4 Manual Tuning vs. Auto Tuning
5 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 DL06 will
6 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
7 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.
8 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
9 consider the impact of any changes to minimize the risk of injury to personnel or damage to
equipment. The auto tune in the DL06 is not intended to be used as a replacement for your
10 process knowledge.

Open-Loop Test
11 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
12 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
13 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
14 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.
A • 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
B 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
C 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.

8–40 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Manual Tuning Procedure

It is not necessary to try to obtain the best values for the P, I and D parameters in the PID 1
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
2
feature in DirectSOFT (see page 8-49).
3
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
4
tuning process.

• Adjust the gains so the Proportional Gain = 0.5 or 1.0 (1.0 is a good value based on experience),
5
Integral Gain = 9999 (this basically eliminates reset) and Derivative Gain = 0000. This disables the
integrator and derivative terms, and provides some proportional gain. 6
• 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. 7
• 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.
8
60% here error 9
10% of Over-damped PV response
10
SP range

SP 50% here 11
PV
12
• 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 13
not desirable since it takes a long time to correct the error; also, there is a difference between
the SP and the PV. 14
60% here
A
10% of B
SP range
PV response
SP 50% here C
PV

D
• 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.

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–41

 Chapter 8: PID Loop Operation

Error
60% here
1
2 50% here
SP
3 PV

4 • 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
5 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,
6 continue to decrease the reset value until the response becomes unstable. See the figure below.

7 60% here

10% of
8 SP range

SP 50% here
9 PV Under-damped PV response

10 • 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
11 destroy the process.
• The control output response should be optimal now, without a Derivative gain. The example
12 recorded values are: Proportional gain = 9.7 and Integral gain = 38 seconds. Note that the error has
been minimized.

13 Minimum Oscillations

14
A Shortest response time

B 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
C 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
D 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.

8–42 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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
1
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. 2
3
4
5
6
7
8
9
10
11
12
13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–43

 Chapter 8: PID Loop Operation

Alternative Manual Tuning Procedures by Others

1 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.
2 Tuning PID Controllers
Two-Mode Simple Method - – for P-I controllers
3 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.
4 3. Make setpoint changes to observe offset (error).
4. Increase reset to eliminate offset (error).
5 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.
6 Zeigler-Nichols Method– “Quarter amplitude decay”
1. Turn off reset and rate; set the proportional gain to a fairly large value.
7 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
8 the ultimate gain (Gu).
4. Measure the period of cycling in minutes. This is the ultimate period (Pu).
9 5. Calculate the controller adjustments as follows:
P only: G = Gu/2
10 P & I : G = Gu/2.2
Ti = 1.2/Pu (repeats/minute)
P-I-D: G = Gu/1.6
11 Ti = 2.0/Pu (repeats/minute)
Td = Pu/8.0 (minutes)
12 Pessen Method
1. Follow the procedure described above (Zeigler-Nichols) to determine the ultimate gain
13 and ultimate period.
2. Apply the formulas below.
14 For no overshoot during startup:
G = Gu/5.0
Ti = 3/Pu (repeats/minute)
A Td = Pu/2 (minutes)
For some overshoot, but better response to disturbances:
B G = Gu/3
Ti = 3/Pu (repeats/minute)
C Td = Pu/3 (minutes)

8–44 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Auto Tuning Procedure

The auto tuning feature for the DL06 loop controller will only run once each time it is 1
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
2
order to derive new gains required for optimal control.
WARNING: Only authorized personnel fully familiar with all aspects of the process should make 3
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
4
damage to equipment. The auto tune in the DL06 is not intended to be used as a replacement for
your process knowledge. 5
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
6
to establish initial PID parameter values. Auto tuning is the best “guess” the CPU can do after
some trial tests. 7
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 8
tuning.
The controls for the auto tuning function use three bits in the PID Mode 2 word V+01, as 9
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
10
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 11
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.
12
Auto Tune Function
13
Start Auto Tune Auto Tune
(0 to 1 transition)
0=PID tuning,
Active
Auto Tune
14
Auto Tuning 1=open PI tuning Error Auto Tuning
Controls 0=closed loop,
1=open loop
Status A
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
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–45

 Chapter 8: PID Loop Operation

Open-Loop Auto Tuning

1 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
2 control output values are in the middle of their ranges (away from the end points).
PLC System
Process Variable
3 Response Step Function
Open Loop

4 Auto Tuning

Control
Setpoint Value Error Term Output
5 + k
–
Loop
Calculation
Manufacturing
Process

Process Variable
6
NOTE: In theory, the SP value does not matter in this case, because the loop is not closed. However,
7 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 DL06, 12 bit PV should be 205 counts or more
8 below the SP for forward-acting loops, or 205 counts or more above the SP for reverse-acting
loops).

9 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
10 the loop table.
The following timing diagram shows the events which occur in the open-loop auto tuning
11 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
12 step change on the output is increased to 20%-of-span.
* When Auto Tune starts, step change output m=10%
13 * 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.

14 * When PV change is under 2%, output is changed at 20%. Open Loop Auto Tune Cycle Wave: Step
PV
Rr = Slope
(%) Tangent
SP
A Process Wave
Base Line
LrRr

B (%)
Lr
(sec.)
Time (sec)

C Output Value
Step Change im=10%

(%)
D PID Cycle
Auto Tune Cycle
PID Cycle

Auto Tune Start Auto Tune End

8–46 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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:
1
PID Tuning
P=1.2*m/LrRr
PI Tuning
P=0.9*m/LrRr
2
I=2.0* Lr I=3.33* Lr
D=0.5* Lr D=0 3
Sample Rate = 0.056* Lr Sample Rate = 0.12*Lr
m = Output step change (10% = 0.1, 20% = 0.2) 4
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 5
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
6
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 7
section on the manual tuning procedure (located prior to this auto tuning section).
Closed-Loop Auto Tuning 8
During a closed-loop auto tuning cycle the loop controller operates as shown in the diagram
below. 9
PLC System
Process Variable
10
Response Limit cycle wave
Closed Loop
Auto Tuning 11
Setpoint V alue
+ k
Error T erm Loop
Calculation
Control
Output Manufacturing
Process
12
–
Process Variable 13
When auto tuning, the loop controller imposes a square wave on the output. Each transition 14
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
A
places the results in the corresponding registers in the loop table.
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–47

 Chapter 8: PID Loop Operation

The following timing diagram shows the events which occur in the closed-loop auto tuning
1 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
2 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.

3 Process Wave
SP
Xo

PV

4 Output Value

5 M

6 To
PID Cycle PID Cycle

7 Auto Tune Cycle

Auto Tune Start Auto Tune End

Calculation of
PID parameter

8
*Mmax = Output Value upper limit setting. Mmin = Output Value lower limit setting.
9 * This example is direct–acting.
When set to reverse–acting, the output will be inverted. When the loop tuning observations
10 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
11 according to the Zeigler-Nichols equations shown below:

12 Kpc = 4M / ( *X0) Tpc = 0

M = Amplitude of output
PID Tuning PI Tuning
13 P = 0.45*Kpc P = 0.30*Kpc
I = 0.60*Tpc I = 1.00*Tpc
14 D = 0.10*Tpc
Sample Rate = 0.014*Tpc
D=0
Sample Rate = 0.03*Tpc

A Auto tuning error

In open-loop tuning, if the auto tune error bit (bit 13 of loop Mode/Alarm status word
B 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.
C 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).
D

8–48 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Use Direct SOFT 5 Data View with PID View

The Data View window is a very useful tool which can be used to help tune your PID loop. 1
You can compare the variables in the PID View with the actual values in the V-memory
location with Data View.
2
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
3
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 4
example of a newly opened Data View. The window will open next to the Ladder View by
default. 5
6
7
8
9
10
11
12
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.
13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–49

 Chapter 8: PID Loop Operation

Open PID View

1 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.
2 The PID View can also be opened by clicking on the PID View button from the PLC Setup
toolbar if it is in view.

3
4
5
6
7
8
9
10 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
11 sized for better use as shown on the facing page.

12
13
14
A
B
C
D

8–50 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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. 1
2
3
4
5
6
7
8
The diagram below illustrates how the 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 8-20 for
details of each word in the table. This is also a good data type reference for each word in the
9
table.
10
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. 11
Process Variable and
12
Setpoint trends are
color coded. 13
The loop name area 14
turns red whenever there
is an overflow error.
A
P
B
I C
D
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–51

 Chapter 8: PID Loop Operation

With both windows positioned in this manner, you are able to see where the PID values have
1 been set and see the process that it is controlling. In the diagram below, you can see the
current SP, PV and Output values, along with the other PID addresses. Refer to the Loop
Table Word Definitions (page 8-20) for details for each word in the table. This is also a good
2 data type reference for each word in the table.

3
4
5
6
7
8
9
10
11
12
13
14
A
B
C
D

8–52 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Using the Special PID Features

It’s a good idea to understand the special features of the DL06 and how to use them. You may
1
want to incorporate some of these features for your PID.
How to Change Loop Modes
2
The first three bits of the PID Mode 1 word (V+00) request
the operating mode of the corresponding loop. Note: these
PID Mode 1 Setting V+00

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
3
bits are mode change requests, not commands (certain
conditions can prohibit a particular mode change – see next
page).
Cascade Manual 4
Automatic
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
5
resets the bits back to “000” after it reads the mode change request. Methods of requesting
mode changes are: 6
• 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. 7
• 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 not
8
use an OUT coil). On a 0–1 transition of X0, the rung sets the Go to Auto Mode
Auto bit equal to 1. The loop controller resets it.
X0 B2000.1
SET
9
• Operator panel – interface the operator’s panel to ladder logic
using standard methods, then use the logic to the right to set the
mode bit.
10
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
11
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 12
possible mode-dependent SP sources, and the two possible Control Output sources.

Input from Operator

Manual
13
Control Output
Cascade
from another loop
Setpoint
+ k Error T erm Loop
Calculation
Control Output
14
Normal Source – Auto/Cascade
Auto/Manual Process Variable
Mode Select
A
PID Mode
Control B
PID Mode 1 Setting V+00 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
C
Mode Request Mode Monitoring
Cascade Manual
Automatic
Cascade Manual
Automatic D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–53

 Chapter 8: PID Loop Operation

Operator Panel Control of PID Modes

1 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
2 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.
3 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
4 to the corresponding data locations.
Operator’s Panel
5 Manual

Auto
6 Mode Request
Cascade
Mode Monitoring

7 PID Mode 1 Setting V+00

Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0
Loop Mode and Alarm Status V+06

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

8 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:
9 • 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
10 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
11 such, the CPU records the modes of all 4 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).
12 • 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.
13 • You can add and configure new loops only when the PLC is in Program Mode. New loops
automatically begin in Manual Mode.
14 Loop Mode Override
In normal conditions the mode of a loop is determined by the request to V+00, bits 0,
A 1, and 2. However, some conditions exist which will prevent a requested mode change
from occurring:
B • 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
C Cascade mode.
In other situations, the PID loop controller will automatically change the mode of the loop to
D 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.

8–54 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

PV Analog Filter
A noisy PV signal can make tuning difficult and can cause the control output to be more 1
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 DL06’s built-in filter. The second method
2
achieves a similar result using ladder logic.
The DL06 Built-in Analog Filter 3
The DL06 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 4
may disable the filter after auto tuning is complete, or continue to use it if the PV input
signal is noisy. 5
+ k
Loop
Calculation
Control Output 6
– Unfiltered
0
PV
7
Process Variable
1 Filtered
PV
8
P ID Mode 2 Setting V+01
Loop Table
9
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

PV filter
enable/disable
V+24 XXXX FIlter constant
10
Bit 2 of PID Mode Setting 2 provides the enable/disable control for the low-pass PV filter 11
(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:
12
• 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.
13
• 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
14
roll-off frequencies.
We highly recommend using DirectSOFT for the auto tuning interface. The duration of each A
auto tuning cycle will depend on the mass of your process. A slowly-changing PV will result
in a longer auto tune cycle time. B
When the auto tuning is complete, the proportional and integral gain values are automatically
updated in loop table locations V+10 and V+11 respectively. The derivative is calculated if
you autotune for PID and updated in loop table location V+12. The sample time in V+07 is
C
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 D
instructions on how to do this are in the section on the manual tuning procedure.

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–55

 Chapter 8: PID Loop Operation

The algorithm which the built-in filter follows is:

1 yi = k (xi – yi–1) + yi–1
yi is the current output of the filter
2 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
3 Creating an Analog Filter in Ladder Logic
4 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
5 application.
Filtering can induce a 1 part in 1000 error in your output because of “rounding.” If your
6 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
7 constant the greater the smoothing effect, but the slower the response time. Be sure a slower
response is acceptable in controlling your process.

8 SP1
LD
Loads the analog signal, which is a BCD value
and has been loaded from V-memory location
V2000 V2000, into the accumulator. Contact SP1 is

9 BIN
always on.
Converts the BCD value in the accumulator
to binary. This instruction is not needed if the
analog value is originally brought in as a

10 BTOR
binary number.

Converts the binary value in the accumulator

to a real number.

11 SUBR
Subtracts the real number stored in location
V1400 from the real number in the
V1400 accumulator, and stores the result in the

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

14 stores the result in the accumulator.

Copies the value in the accumulator

OUTD to location V1400.
V1400
A
Maintenance

Converts the real number in the

RTOB accumulator to a binary value, and
stores the result in the accumulator.

B 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

C binary number).

Loads the BCD number filtered value from

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

8–56 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Use the DirectSOFT 5 Filter Intelligent Box Instruction

For those who are using DirectSOFT 5, you have the opportunity to use Intelligent Box 1
(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
2
New =New Filtered Value
Old = Old Filtered Value
3
FDC = Filter Divisor Constant
Raw = Raw Data
4
The Filter Divisor Constant is an integer in
the range K1 to K100, such that if it equaled
5
K1, then no filtering is performed.
The rate at which the calculation is performed is specified by time in hundredths of a second
6
(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 7
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.
8
Since the following binary filter example does not write directly to the PID PV location, the
BCD filter could be used with BCD values and then converted to BIN.
9
FilterB Example
Following is an example of how the FilterB IBox is used in a ladder program. The instruction
10
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 11
will increase the smoothing effect of the filter. A value of 1 results with no filtering. The
filtered value will be placed in V2100. 12
13
14
A
NOTE: See Chapter 5, page 242, for more detailed information.
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–57

 Chapter 8: PID Loop Operation

1 Ramp/Soak Generator
Introduction
2 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
3 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.
4 Setpoint Sources:
Operator Input
5 Ramp/soak generator
Ladder Program
Setpoint V+02
k +
–
Loop
Calculation
Control Output

Another loop’s output (cascade)

6 Process Variable

If the SP for your process rarely changes or can tolerate step changes, you probably will not
7 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
8 for these applications.
The terms ramp and soak have special meanings in the SP
9 process control industry, and refer to desired setpoint (SP)
values in temperature control applications. In the figure to Soak
Ramp
the right, the setpoint increases during the ramp segment.
10 It remains steady at one value during the soak segment. slope
Complex SP profiles can be generated by specifying a series
11 of ramp/soak segments. The ramp segments are specified in
Time

SP units per second time. The soak time is also programmable in minutes.
12 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
13 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
14 start/stop functionality of the ramp/soak generator. The ladder program can monitor the
status of the ramp soak profile (current ramp/segment number).
A Ramp/soak table
Ramp/soak Setpoint Loop Control Output
k
B Ramp/soak controls Generator +
–
Calculation

Process Variable
C
D

8–58 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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).
1
• You may specify up to eight ramp/soak steps (16 segments). 2
• 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.
3
• Ramp/soak monitoring includes Profile Complete, Soak Deviation (SP minus PV), and current
ramp/soak step number. 4
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 5
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.
6
15 16 7
13 14 Soak

5 6
Ramp
Soak
Ramp
8
3 4 Soak
Step 1 2
Ramp
Soak
Ramp
9
Soak
Ramp
SP 10
Ramp/Soak Table V–Memory Space
11
The parameters which define the ramp/soak
profile for a loop are in a ramp/soak table. Each User Data 12
loop may have its own ramp/soak table, but it is
optional. Recall the Loop Parameter table consists
of a 32-word block of memory for each loop, and
V2000
V2037
V2040
LOOP #1
32 words
V2034 =
3000 octal 13
LOOP #2
together they occupy one contiguous memory V2074 =
area. However, the ramp/soak table for a loop is
V2077 32 words
3600 octal 14
individually located, because it is optional for each
loop. An address pointer in location V+34 in the
loop table specifies the starting location of the
V3000 Ramp/Soak #1
32 words
A
ramp/soak table.
In the example to the right, the loop parameter
B
tables for Loop #1 and #2 occupy contiguous 32-
word blocks as shown. Each has a pointer to its V3600 Ramp/Soak #2
32 words
C
ramp/soak table, independently located elsewhere
in user V-memory. Of course, you may locate all
the tables in one group, as long as they do not
D
overlap.

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–59

 Chapter 8: PID Loop Operation

The parameters in the ramp/soak table must be user-defined. the most convenient way is to
1 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.
2 • 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).
3 • 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).
4 • Soak Duration – specifies the time for the soak segment in minutes, ranging from 000.1 to 999.9
minutes in BCD (implied decimal point).

5 • 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.

6 Ramp End
SP Value Ramp/Soak Table
Soak PV

7 deviation V+00
V+01
XXXX
XXXX
Ramp End SP Value
Ramp Slope
Slope Soak
SP V+02 XXXX Soak Duration
8 duration
segment becomes active
V+03 XXXX Soak PV Deviation

9 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
10 assumes the present SP as the starting SP.

Offset Step Description Offset Step Description

11 + 00 1 Ramp End SP Value + 20 9 Ramp End SP Value
+ 01 1 Ramp Slope + 21 9 Ramp Slope
12 + 02
+ 03
2
2
Soak Duration
Soak PV Deviation
+ 22
+ 23
10
10
Soak Duration
Soak PV Deviation
+ 04 3 Ramp End SP Value + 24 11 Ramp End SP Value
13 + 05 3 Ramp Slope + 25 11 Ramp Slope
+ 06 4 Soak Duration + 26 12 Soak Duration
14 + 07
+ 10
4
5
Soak PV Deviation
Ramp End SP Value
+ 27
+ 30
12
13
Soak PV Deviation
Ramp End SP Value
+ 11 5 Ramp Slope + 31 13 Ramp Slope
A + 12 6 Soak Duration + 32 14 Soak Duration
+ 13 6 Soak PV Deviation + 33 14 Soak PV Deviation
B + 14
+ 15
7
7
Ramp End SP Value
Ramp Slope
+ 34
+ 35
15
15
Ramp End SP Value
Ramp Slope
+ 16 8 Soak Duration + 36 16 Soak Duration
C + 17 8 Soak PV Deviation + 37 16 Soak PV Deviation

8–60 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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. 1
Ramp/Soak Table Flags
The individual bit definitions of the Ramp/Soak Table Flag (Addr+33) word is listed in the
following table.
2
Bit Ramp/Soak Flag Bit Description Read/Write Bit=0 Bit=1 3
0 Start Ramp / Soak Profile write – 01 Start
1
2
Hold Ramp / Soak Profile
Resume Ramp / Soak Profile
write
write
–
–
01 Hold
01 Resume
4
3 Jog Ramp / Soak Profile write – 01 Jog
4
5
Ramp / Soak Profile Complete
PV Input Ramp / Soak Deviation
read
read
–
Off
Complete
On
5
6 Ramp / Soak Profile in Hold read Off On
7 Reserved read Off On 6
8–15 Current Step in R/S Profile read decode as byte (hex)

Ramp/Soak Generator Enable

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

to the right. The other ramp/soak controls in V+33

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

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–61

 Chapter 8: PID Loop Operation

The normal state for the ramp/soak control bits is all zeros. Ladder logic must set only one
1 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
2 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
3 be frozen.
• Resume – a 0 to 1 transition cause the ramp/soak generator to resume operation if it is in the hold
4 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),
5 and go to the next segment.
Ramp/Soak Profile Monitoring
6 You can monitor the Ramp/Soak profile status using other bits
in the Ramp/Soak Settings V+33 word, shown to the right.
Ramp/Soak Settings V+33

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

7 • R/S Profile Complete – =1 when the last programmed step is

done.
R/S Profile in Hold
Soak PV Deviation
R/S Profile Complete
• Soak PV Deviation – =1 when the error (SP–PV) exceeds the
8 specified deviation in the R/S table.
• R/S Profile in Hold – =1 when the profile was active but is now
9 in hold. Ramp/Soak Settings V+33
The number of the current step is available in the upper 8 bits Ramp/Soak Settings V+33

10 of the Ramp/Soak Settings V+33 word. The bits represent a 2-

digit hex number, ranging from 1 to 10. Ladder logic can
Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

monitor these to synchronize other parts of the program with Current Profile Step, 2–digit hex
11 the ramp/soak profile. Load this word to the accumulator and Value = 01 to 10 hex,
or 1 to 16 decimal
shift right 8 bits, and you have the step number.
12 Ramp/Soak Programming Errors
The starting address for the ramp/soak table must be a valid Ramp/Soak Table Error V+35

13 location. If the address points outside the range of user V- Bit 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1 0

memory, one of the bits to the right will turn on when the
Starting Address set in
14 ramp/soak generator is started. We recommend using
DirectSOFT to configure the ramp/soak table. It automatically
reserved system V-memory
Starting Address set out of
V-memory upper range
range checks the addresses for you.
A Testing Your Ramp/Soak Profile
Starting Address set out
of V-memory lower range

It’s a good idea to test your ramp/soak profile before using it to control the process. This is
B 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-
C screen for you. Be sure to set the trending timebase slow enough to display completed ramp-
soak segment pairs in the waveform window.
D

8–62 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

DirectSOFT Ramp/Soak Example

The following example will step you through the Ramp/Soak setup.
1
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
2
Setup PID window and select the R/S tab, and then enter the Ramp and Soak data.
3
4
5
6
7
8
Program the Ramp/Soak Control in Relay Ladder 9
Refer to the Ramp/Soak Flag Bit Description table on page 8-60 when adding the control
rungs to your program similar to the ladder rungs below. 10
11
12
13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–63

 Chapter 8: PID Loop Operation

Test the Profile

1 Test your profile using PID View.

2
3
4
5
6
7
8
9
10
11
12
13
14
A
B
C
D

8–64 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Cascade Control
1
Introduction
Using cascaded loops is an advanced control technique, superior to individual loop control in 2
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 DL06 also
provides Cascaded Mode.
3
NOTE: Using cascaded loops is an advanced process control technique; therefore, we recommend
their use only for experienced process control engineers.
4
When a manufacturing process is complex and contains a lag time from control input to 5
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.
6
Identifying the intermediate variable allows us to divide the process into two parts as shown
in 7
PROCESS the
Control input Process A
Intermediate
Variable Process B
Process
Variable (PV) 8
following figure. 9
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. 10
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 11
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
12
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 13
process variable at the system level. The cascaded loops behave as one loop, but with
External
Disturbances
External
Disturbances
14
Output B/
Setpoint
+ k
Loop B
Calculation
Setpoint A
+ k
Loop A
Calculation
Output A Process A
(secondary)
Process B
(primary)
A
– –

Major
Minor
Loop
B
Loop PV, Process A
PV, Process B
C
improved performance over the previous single-loop solution.
One of the benefits to cascade control can be seen by examining its response to external D
disturbances. Remember the minor loop is faster acting than the major loop. Therefore, if a

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–65

 Chapter 8: PID Loop Operation

Cascaded Loops in the DL06 CPU

1 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
2 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.

3 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.
4
You can cascade together as many loops as necessary on the DL06, and you may have
5 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.
6 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
7 minor loop will ignore the its local SP register (V+02), and read the major loop’s control
output as its SP instead.
8 Major Loop (Auto mode) Minor Loop (Cascade Mode)
Loop Table Loop Table

9 V+02 XXXX SP V+02 XXXX SP

V+03 XXXX PV V+03 XXXX PV

10 V+05 XXXX Control Output V+05 XXXX Control Output

11 V+32 XXXX Remote SP Pointer

When using DirectSOFT’s PID View to watch the SP value of the minor loop, DirectSOFT
12 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.
13 Now, we use the loop parameter arrangement above and draw its equivalent loop diagram,
shown below.
14
Major loop Minor Cascaded loop
A Loop Control Output V+05 Cascade
Control
Calculation
Remote Setpoint Loop Output
B SP + k Calculation
Local SP
–
V+02 Auto/Manual Process Variable
C
D Remember that a major loop goes to Manual Mode automatically if its minor loop is taken
out of Cascade Mode.

8–66 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Tuning Cascaded Loops

In tuning cascaded loops, you will need to de-couple the cascade relationship and tune the 1
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
2
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 3
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
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 its
5
overall process. Therefore, do not go back and tune the minor loop again while tuning the major
loop. 6
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. 7
8
9
10
11
12
13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–67

 Chapter 8: PID Loop Operation

1 Time-Proportioning Control
The PID loop controller in the DL06 CPU generates a smooth control output signal across a
numerical range. The control output value is suitable to drive an analog output module,
2 which connects to the process. In the process control field, this is called continuous control,
because the output is on (at some level) continuously.
3 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-
4 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.
5 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
6
7
8
9
10
11 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
12 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
13 the process variable.
Time-proportioning control approximates continuous control by virtue of its duty-cycle – the
14 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.
A period

Desired

B Effect

On/Off On
Control Off
C 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.
D

8–68 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

On/Off Control Program Example

The following ladder segment provides a time proportioned on/off control output. It converts 1
the continuous output in V2005 to on/off control using the output coil, Y0.

SP Loop V2005 Time Y0 P

2
+ k Calculation Proportioning
Process
V
–
PV
continuous on/off
3
The example program uses two timers to generate On/Off control. It makes the following
assumptions, which you can alter to fit your application:
4
• 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).
5
• 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).
6
• The On/Off control output is Y0.
The time proportioning program must match the resolution of the output (1 part in 1000) to
7
the resolution of the time base of T0 (also 1 part in 1000).
8
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
9
instead of a relay.
T0 A fast timer (0.01 sec. timebase) establishes the primary
10
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. 11
T0 At the end of the 10 second period, T0 turns on, and
LD
V2005 loads the control output value (binary) from the loop table
V+05 location (V2005).
The BTOR instruction changes the number in the
12
BTOR
accumulator to a real number.

DIVR
R4.095
Dividing the control output by 4.095, converts the
0 – 4095 range to 0 – 1000, which “matchs” the
13
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.
14
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

A
OUT location of the timer preset for the second timer.

T0
V1400
The second fast timer also counts in increments of .01
B
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,

T1 TA1 K0
turns off the output coil, Y0, when the preset is reached.

The N.C. T1 contact, inverts the T1 timer output. The

C
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.
D
END END coil marks the end of the main program.

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–69

 Chapter 8: PID Loop Operation

1 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
2 setpoint. Use of this feature is an option available to you on the DL06. However, it’s best to
implement and tune a loop without feedforward, and adding it only if better loop
3 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.
4 Feedforward path
kf
5
+
6 Setpoint
k
Loop
Calculation
+ k
Control Output
+
–

7 Process Variable
In the previous section on the bias term, we said that “the bias term value establishes a
8 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
9 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
10 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
11 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
12 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.
13 • Proper use of feedforward will allow us to reduce the integrator gain. Reducing integrator gain gives
us an even more stable control system.
14 Feedforward is very easy to use in the DL06 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
A location V+04.

B Parameter Table location V+04.

Loop Calculation

C kp P

V+04
Setpoint Error T erm + Control Output
k ki XXXX k
Bias Term
D +
–
I +
+

Process Variable kd D

8–70 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

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
1
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 2
example).
NOTE: When writing the bias term, one must be careful to design ladder logic to write the value only
3
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. 4
Feedforward Example
How do we know when to write to the bias term, and what value to write? Suppose we have
5
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 6
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.
7
Oven Closed
Open Closed
8
door

PV PV sags
9
PV excess

10
Bias
11
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 12
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
13
shows the results.

Oven Closed
14
Open Closed
door
A
PV

Feed-forward Feed-forward
B
Bias
C
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
D
in setpoint.

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–71

 Chapter 8: PID Loop Operation

1 PID Example Program

Program Setup for the PID Loop
2 After setting up the PID loop or loops, 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
3 loop(s).
The following example program shows how an RTD module, F0-04RTD, and an analog
4 combination module, F0-2AD2DA-2, are used and setup for a PID loop. This example
assumes that the PID table for loop 1 has a beginning address of V2100.
5 All of the analog I/O modules used with the DL06 are setup in a similar manner. Refer to the
DL05/DL06 Options Manual for the setup information for the particular module that you
will be using.
6 DirectSOFT

7
8
Note: The inputs will be read in binary format 0-4095 only if the temperature does not
9 go above 409.5 degrees. Full range is 65535, which equates to 6,553.5 degrees, since the
RTD card reads directly in tenths of a degree. The input resolution of the PID loop needs
to be set based on the max. temperature of the application since the RTD card is always
10 16 bit resolution.

11
12
13
14 Set to read degrees F and select up scale burnout.

A
B
C
D
Program continued on next page

8–72 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Example program continued

1
2
3
4
5
6
7
8
9
10
11
12
Note: The value stored in V1400 must be in the same scale as
the PV value, or tenths of a degree in this example. 13
14
Manual Mode Request
B2100.0 A
Auto Mode Request
B2100.1
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–73

 Chapter 8: PID Loop Operation

Note that the modules used in the PID loop example program were set up for binary format.
1 They could have been set up for BCD format. In the latter 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 BCD to binary before being stored to
2 the analog output.
By following the steps outlined in this chapter, you should be able to set up workable PID
3 control loops. The DirectSOFT Programming Software Manual provides more information
for the use of PID View.
4 For a step-by-step tutorial, go to the Technical Support section located on our website,
www.automationdirect.com. Once you are at the website, click on Technical Support Home.
5 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
6 Powerpoint Viewer can be downloaded if your computer does not have Powerpoint installed.

7
8
9
10
11
12
13
14
A
B
C
D

8–74 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Troubleshooting Tips
1
Q. The loop will not go into Automatic Mode.
A. Check the following for possible causes: 2
• 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. 3
Q. The Control Output stays at zero constantly when the loop is in Automatic
Mode. 4
A. Check the following for possible causes:
• The Control Output upper limit in loop table location V+31 is zero. 5
• The loop is driven into saturation, because the error never goes to zero value and changes (algebraic)
sign. 6
Q. The Control Output value is not zero, but it is incorrect.
A. Check the following for possible causes: 7
• 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.
8
Q. The Ramp/Soak Generator does not operate when I activate the Start bit. 9
A. Check the following for possible causes:
• 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.
10
• 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
11
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. 12
• 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. 13
• 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.
14
Q. The PV value in the table is constant, even though the analog module receives A
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
B
the ladder is working.
Q. The Derivative gain doesn’t seem to have any affect on the output.
C
A. The derivative limit is probably enabled (see section on derivative gain limiting). D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–75

 Chapter 8: PID Loop Operation

Q. The loop Setpoint appears to be changing by itself.

1 A. Check the following for possible causes:
• The Ramp/Soak generator is enabled, and is generating setpoints.
2 • 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
3 quick way to do this is to temporarily place an end coil at the beginning of your program, then go
to PLC Run Mode.

4 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.
5 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
6 these in hex, so DirectSOFT converts them for you. The values in the table range from 0
to FFF, for 12-bit unipolar format.
7 Q. The loop seems unstable and impossible to tune, no matter what gains I use.
A. Check the following for possible causes:
8 • The loop sample time is set too long. Refer to the section near the front of this chapter on selecting
the loop update time.
9 • 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.

10 • 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.
11 • 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.
12
13
14
A
B
C
D

8–76 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Glossary of PID Loop Terminology

Automatic Mode An operational mode of a loop, in which it makes PID calculations and updates the
1
loop’s control output.
Bias Freeze A method of preserving the bias value (operating point) for a control output, by inhibiting
2
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 3
control output value.
Bumpless Transfer A method of changing the operation mode of a loop while avoiding the usual 4
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. 5
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.
6
Cascade Mode An operational mode of a loop, in which it receives its SP from another loop’s output.
Continuous Control Control of a process done by delivering a smooth (analog) signal as the control
output.
7
Control Output The numerical result of a PID equation which is sent by the loop with the intention of
nulling out the current error.
8
Derivative Gain A constant that determines the magnitude of the PID derivative term in response to
the current error.
9
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. 10
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.
11
You can specify the size of the deadband.
Error Squared An optional feature which multiplies the error by itself, but retains the original algebraic
12
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 13
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 14
current error.
Major Loop In cascade control, it is the loop that generates a setpoint for the cascaded loop. A
Manual Mode An operational mode of a loop, it which the PID calculations are stopped. The operator
must manually control the loop by writing to the control output value directly. B
Minor Loop In cascade control, the minor loop is the subordinate loop that receives its SP from the
major loop. C
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 DL06’s continuous loop output to on/off control.
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–77

 Chapter 8: PID Loop Operation

PID Loop A mathematical method of closed-loop control involving the sum of three terms based on
1 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.

2 Position Algorithm The control output is calculated so it responds to the displacement (position) of the
PV from the SP (error term)

3 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,
4 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
5 to the current error.
PV Absolute Alarm A programmable alarm that compares the PV value to alarm threshold values.
6 PV Deviation Alarm A programmable alarm that compares the difference between the SP and PV
values to a deviation threshold value.
7 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
8 of programming the PLC to generate such SP sequences.
Rate Also called differentiator, the rate term responds to the changes in the error term.
9 Remote Setpoint The location where a loop reads its setpoint when it is configured as the minor loop in
a cascaded loop topology.
10 Reset Also called integrator, the reset term adds each sampled error to the previous, maintaining a
running total called the bias.
11 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.
12 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.
13 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.
14 Setpoint (SP) The desired value for the process variable. The setpoint (SP) is the input command to the
loop controller during closed loop operation.
A 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.
B 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)
C 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,
D 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.

8–78 DL06 Micro PLC User Manual, 3rd Edition, Rev. C

 Chapter 8: PID Loop Operation

Bibliography
1
Fundamentals of Process Control Theory, Second Edition Application Concepts of Process Control
Author: Paul W. Murrill
Publisher: Instrument Society of America
Author: Paul W. Murrill
Publisher: Instrument Society of America 2
ISBN 1–55617–297–4 ISBN 1–55617–080–7
PID Controllers: Theory, Design, and Tuning, 2nd Edition Author:
K. Astrom and T Hagglund
Fundamentals of Temperature, Pressure, and Flow Measurements,
Third edition 3
Publisher: Instrument Society of America Author: Robert P. Benedict
ISBN 1–55617–516–7 Publisher: John Wiley and Sons

Process / Industrial Instruments & Controls Handbook, Fourth

ISBN 0–471–89383–8
pH Measurement and Control, Second Edition
4
Edition Author: Gregory K. McMillan
Author (Editor-in-Chief): Douglas M. Considine
Publisher: McGraw-Hill, Inc ISBN 0-07-012445-0
Publisher: Instrument Society of America
ISBN 1–55617–483–7
5
Programmable Controllers Concepts and Applications, First Fundamentals of Programmable Logic Controllers, Sensors, and
Edition
Authors: C.T. Jones and L.A. Bryan
Communications
Author: Jon Stenerson
6
Publisher: International Programmable Controls Publisher: Prentice Hall ISBN 0-13-726860-2
ISBN 0-915425-00-9
Process Control, Third Edition Instrument Engineer’s Handbook Process Measurement and Analysis, Third Edition Instrument
7
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 8
9
10
11
12
13
14
A
B
C
D

DL06 Micro PLC User Manual, 3rd Edition, Rev. C 8–79