EMERSON PACSystems VersaMax MICRO PLC & NANO PLC. USER MANUAL (2019)

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

562

Parameter Low Bit Units Range

Description

Address+5

Proportional
Gain -Kp
(Controller
gain, Kc, in
the ISA
version)

0.01 CV%/PV%

0 to 327.67%

Change in the control variable in CV Counts for a 100
PV Count change in the Error term. A Kp entered as
450 is displayed as 4.50 and results in a Kp*Error/100
or 450*Error/100 contribution to the PID Output. Kp
is generally the first gain set when adjusting a PID loop.

Address+6

Derivative
Gain-Kd

0.01 seconds

0 to 327.67 sec Change in the control variable in CV Counts if the Error

or PV changes  1 PV Count every  10ms. Entered as a
time with the low bit indicating 10ms. For example, a
Kd entered as 120 is displayed as 1.20 Sec and results
in  a  Kd  *  delta  Error/delta  time  or  120*4/3
contribution to the PID Output if Error was changing
by 4 PV Counts every 30ms. Kd can be used to speed
up  a  slow  loop  response  but  is  very  sensitive  to  PV
input noise.

Address+7

Integral Rate-
Ki

Repeat/1000
Sec

0 to 32.767
repeat/sec

Change in the control variable in CV Counts if the Error
were  a  constant  1  PV  Count.    Displayed  as  0.000
Repeats/Sec  with  an  implied  decimal  point  of  3.  For
example,  a  Ki  entered  as  1400  is  displayed  as  1.400
Repeats/Sec and results in a Ki * Error * dt or 1400 *
20 * 50/1000 contribution to PID Output for an Error
of 20 PV Counts and a 50ms PLC sweep time (Sample
Period of 0). Ki is usually the second gain set after Kp.

Address+8

CV Bias/
Output
Offset

CV Counts

-32000 to
32000 (add to
integrator
output)

Number of CV Counts added to the PID Output before
the rate and amplitude clamps.  It can be used to set
non-zero  CV  values  if  only  Kp Proportional  gains  are
used,  or  for  feed  forward  control  of  this  PID  loop
output from another control loop.

Address+9
Address+10

CV Upper
and
Lower
Clamps

CV Counts

-32000 to
32000
(>%Ref+10)

Number  of  CV  Counts  that  define  the  highest  and
lowest  value  for  CV.   These values  are  required.  The
Upper Clamp must have a more positive value than the
Lower Clamp, or the PID block will not work.  These are
usually used to define limits based on physical limits
for  a  CV  output.  They  are  also  used  to  scale  the  Bar
Graph display for CV. The block has anti-reset windup
to  modify  the  integrator  value  when  a  CV  clamp  is
reached.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

563

Parameter Low Bit Units Range

Description

Address+11 Minimum

Slew Time

Seconds/Full
Travel

0 (none) to
32000 sec to
move 32000 CV

Minimum  number  of  seconds  for  the  CV  output  to
move from 0 to full travel of 100% or 32000 CV Counts.
It is an inverse rate limit on how fast the CV output can
be changed.
If  positive,  CV  cannot  change  more  than  32000  CV
Counts  times  Delta  Time  (seconds)  divided  by
Minimum  Slew  Time.  For  example,  if  the  Sample
Period is 2.5 seconds and the Minimum Slew Time is
500  seconds,  CV  cannot  change  more  than
32000*2.5/500  or  160  CV  Counts  per  PID  solution.
The integrator value is adjusted if the CV rate limit is
exceeded. If Minimum Slew Time is 0, there is no CV
rate limit. Set Minimum Slew Time to 0 while tuning or
adjusting PID loop gains.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

564

Parameter Low Bit Units Range

Description

Address+12 Config Word Low 6 bits used

The low 6 bits of this word are used to modify six
default PID settings. The 10 high-order bits should be
set to 0.
Bit 0: Error Polarity. Setting this bit to 1 modifies the
standard PID Error Term from the normal (SP

–

PV) to

(PV

–

SP), reversing the sign of the feedback term. This

is for Reverse Acting controls where the CV must go
down when the PV goes up.
Bit 1: Output Polarity: Setting this bit to 1 inverts the
Output  Polarity  so  that  CV  is  the  negative  of  the
output rather than the normal positive value.
Bit 2: Setting this bit to 1 removes the setpoint from
the Derivative calculation.
Bit  3:  Deadband  Action:  When  this  bit  is  0,  no
deadband  action  is  chosen.  If  the  error  is  within  the
deadband limits, the error is to be zero. Otherwise the
error is not affected by the deadband limit. If  this bit is
1, the deadband action is chosen. If the error is within
the deadband limits, the error is forced to be zero. If,
however, the error is outside the deadband limits, the
error is reduced by the deadband limit (error = error

–

deadband limit).
Bit 4: Anti-reset windup action. When this bit is 0, the
anti-reset windup action uses a reset back calculation.
When  the  output  is  clamped,  this  replaces  the
accumulated Y remainder value with whatever value is
necessary to produce the clamped output exactly.
When the  bit is 1,  this replaces  accumulated Y  term
with  the  value  of  the  Y  term  at  the  start  of  the
calculation. In this way, the pre-clamp Y value is held
as long as the output is clamped.
Bit 5: Enable derivative filtering. When this bit is set to
0, no filtering is applied to the derivative term.
When bit 5 is set to 1, a first order filter is applied. This
limits  the  effects  of  higher  frequency  process
disturbances on the derivative term.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

565

Parameter Low Bit Units Range

Description

Bit 0

–

Bit 5: Remember that the bits are set in

powers of 2.
To set bit 0, add 1 to the Config Word parameter
value.
To set bit 1, add 2.
To set bit 2, add 4.
To set bit 3, add 8.
To set bit 4, add 16.
To set bit 5, add 32.
For example, set Config Word to 0 for default PID
configuration. Add 1 to change the Error Term from
SP - PV to PV - SP, add 2 to change the Output Polarity
from CV = PID Output to CV =

–

PID Output, or add 4

to change Derivative Action from Error rate of change
to PV rate of change, etc.

Address+13 Manual

Command

CV Counts

Tracks CV in
Auto or Sets CV
in Manual

Set to the current CV output while the PID block is in
Automatic mode. When the block is switched to
Manual mode, this value is used to set the CV output
and the internal value of the integrator within the
Upper and Lower Clamp and Slew Time limits.

Address+14 Control Word Maintained by

the PLC, unless
Bit 1 is set.

PLC maintained
unless set
otherwise: low
bit sets Override
if 1.

If the Override low bit is set to 1, this word and other
internal SP, PV and CV parameters must be used for
remote operation of this PID block (see below). This
allows remote operator interface devices, such as a
computer, to take control away from the PLC
program.

CAUTION

If you do not want this to happen, make sure
the Control Word is set to 0. If the low bit is 0,
the next 4 bits can be read to track the status
of the PID input contacts as long as the PID
Enable contact has power.

A discrete data structure with the first five bit positions
in the following format:

Bit Word

Value

Function Status or External

Action if Override bit
set to 1

Override

If 0, monitor block
contacts below. If 1, set
them externally.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

566

Parameter Low Bit Units Range

Description

Manual
/Auto

If 1, block is in Manual
mode; other numbers it
is in Automatic mode.

Enable

Should normally be 1;
otherwise block is never
called.

UP
/Raise

If 1 and Manual (Bit 1) is
1, CV is being
incremented every
solution.

DN
/Lower

If 1 and Manual (Bit 1) is
1, CV is being
incremented every
solution.

Address+15 Internal SP

Set and
maintained by
the PLC

Non-
configurable

Tracks SP in; must be set externally if
Override = 1.

Address+16 Internal CV

“

Tracks CV out.

Address+17 Internal PV

“

Tracks PV in; must be set externally if
Override bit = 1.

Address+18 Output

“

Signed  word  value  representing  the  output  of  the
function  block  before  the  optional  inversion.    If  no
output inversion is configured and the output polarity
bit in the control word is set to 0, this value equals the
CV  output.      If  inversion  is  selected  and  the  output
polarity bit is set to 1, this value equals the negative of
the CV output.

Address+19 Diff Term

Storage

Address+20
Address+21

Int Term
Storage

Used internally for storage of intermediate values. Do
not write to these locations

Address+22 Slew Term

Storage

Address+23
to
Address+25

Clock

Internal elapsed time storage (time last PID executed).
Do not write to these locations

Address+26 Y Remainder

Storage

Holds remainder for integrator division scaling for 0
steady state error.

Address+27
Address+28

SP, PV Lower
and Upper
Range

PV Counts

-32000 to
32000

Optional INT values in PV Counts that define high and
low display values. (Ref +27 must be lower than
Ref+28)

Address+29
to
Address+39

Reserved

N/A

Non-
configurable

29-34 are reserved for internal use; 35-39 are reserved
for external use. Do not use these references.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

567

26.4

PID Algorithm Selection (PIDISA or PIDIND)
and Gains

The PID block can be programmed selecting either the Independent (PID_IND) term or
standard ISA (PID_ISA) versions of the PID algorithm. The only difference in the algorithms
is how the Integral and Derivative gains are defined.

26.4.1

Error Term

The Independent (PID_IND) term and standard ISA (PID_ISA) versions of the PID algorithm
both calculate the Error Term as:

Error Term (in Normal Mode) =

Set Point (SP)

–

Process Variable (PV)

Error Term in Reverse-Action Mode

The Error Term calculation can be changed to Reverse Action mode:

Error Term (in Reverse Action Mode) =

Process Variable (PV)

–

Set Point (SP)

by setting the Error Term (low bit 0 in the Config Word %Ref+12) to 1.

Effect of the Process Variable on the Control Variable

In Normal mode, the Control Variable (CV) responds to changes in the Process Variable (PV)
by moving in the same direction. In Reverse-Action mode, when the Process Variable goes
up, the Control Variable goes down.

26.4.2

Derivative Term

The Derivative term is the time rate of change of the Error term in the interval since the last
PID solution.

Derivative =

∆

Error / Delta time (dt) = (Error

–

previous Error) / Delta time (dt)

where

Delta time (dt) = Current PLC elapsed time - PLC elapsed time at previous PID solution.

In Normal Mode (without Reverse-Action mode), this is equal to the change in the Error
term.

(Error

–

previous Error) = (SP

–

PV)

–

(previous SP

–

previous PV)

(previous PV

–

PV)

–

(previous SP

–

SP)

However, when the Reverse-Action mode bit (bit 0) in the Config Word is set, the sign of the
change in the Error term is reversed.

(Error

–

previous Error) = (PV

–

SP)

–

(previous PV

–

previous SP)

(PV

–

previous PV)

–

(SP

–

previous SP)

The change in the Error term depends on changes in both the Set Point and the Process
Variable. If the Set Point is constant, the difference between SP and the previous SP is zero

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

568

and has no effect on the output. However, Set Point changes can cause large changes in the
Derivative term and in the output.

Removing Set Point Changes from the Derivative Calculation

Loop  stability  can  be  improved  by  eliminating  the  effect  of  Set  Point  changes  on  the
Derivative term.  Set the third bit (bit 2) of the Config Word to 1 to calculate the derivative
based only on the change in PV.

For bit 2 set in normal mode (bit 0 = 0),

(Error

–

previous Error) = (previous PV

–

PV),

and with bit 2 set in Reverse-Action mode (bit 0 = 1),

(Error

–

previous Error) = (PV

–

previous PV).

26.4.3

Delta Time

The dt (or Delta Time) is determined by subtracting the last PID solution clock time for this
block from the current PLC elapsed time clock.

dt =

Current PLC Elapsed Time clock - PLC Elapsed Time Clock at Last PID
solution

Derivative = (Error - previous Error)/dt or

(PV - previous PV)/dt if 3rd bit of Config Word set to 1

The Independent term PID (PID_IND) algorithm calculates the output as:

PID Output = Kp * Error + Ki * Error * dt + Kd * Derivative + CV Bias

The standard ISA (PID_ISA) algorithm has a different form:

PID Output = Kc * (Error + Error * dt/Ti + Td * Derivative) + CV Bias

where Kc is the controller gain, and Ti is the Integral time and Td is the Derivative time. The
advantage of ISA is that adjusting the Kc changes the contribution for the integral and
derivative terms as well as the proportional one, which may make loop tuning easier. If you
have PID gains in terms or Ti and Td, use

Kp = Kc Ki = Kc/Ti and Kd = Kc/Td

to convert them to use as PID User Parameter inputs.

The CV Bias term above is an additive term separate from the PID components. It may be
required if you are using only Proportional Kp gain and you want the CV to be a non-zero
value when the PV equals the SP and the Error is 0. In this case, set the CV Bias to the desired
CV when the PV is at the SP. CV Bias can also be used for feed forward control where another
PID loop or control algorithm is used to adjust the CV output of this PID loop.

If an Integral Ki gain is used, the CV Bias would normally be 0 as the integrator acts as an
automatic bias. Just start up in Manual mode and use the Manual Command word (%Ref+13)
to set the integrator to the desired CV, then switch to Automatic mode. This also works if Ki

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

569

is 0, except the integrator will not be adjusted based on the Error after going into Automatic
mode.

26.4.4

Independent Term Algorithm (PIDIND)

The following diagram shows how the PID algorithms work:

Figure 379

The ISA Algorithm (PIDISA) is similar except the Kp gain is factored out of Ki and Kd so that
the integral gain is Kp * Ki and derivative gain is Kp * Kd. The Error sign, DerivAction and
Polarity are set by bits in the Config Word user parameter.

CV Amplitude and Rate Limits

The block does not send the calculated PID Output directly to CV. Both PID algorithms can
impose amplitude and rate of change limits on the output Control Variable. The maximum
rate of change is determined by dividing the maximum 100% CV value (32000) by the
Minimum Slew Time, if specified as greater than 0. For example, if the Minimum Slew Time
is 100 seconds, the rate limit will be 320 CV counts per second. If the dt solution time was
50 ms, the new CV output cannot change more than 320*50/1000 or 16 CV counts from
the previous CV output.

The CV output is then compared to the CV Upper and CV Lower Clamp values. If either limit
is exceeded, the CV output is set to the clamped value. If either rate or amplitude limits are
exceeded modifying CV, the internal integrator value is adjusted to match the limited value
to avoid reset windup.

Finally, the block checks the Output Polarity (2nd bit of the Config Word %Ref+12) and
changes the sign of the output if the bit is 1.

CV = Clamped PID Output or

- Clamped PID Output if Output Polarity bit set

If the block is in Automatic mode, the final CV is placed in the Manual Command %Ref+13.
If  the  block  is  in  Manual  mode,  the  PID  equation  is  skipped  as  CV  is  set  by  the  Manual
Command,  but  all  the  rate  and  amplitude  limits  are  still  checked.  That  means  that  the
Manual Command cannot change the output above the CV Upper Clamp or below the CV
Lower Clamps and the output cannot change faster than the Minimum Slew Time allowed.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

570

26.4.5

Sample Period and PID Block Scheduling

The PID block is a digital implementation of an analog control function, so the dt sample
time in the PID Output equation is not the infinitesimally small sample time available with
analog controls. The majority of processes being controlled can be approximated as a gain
with a first or second order lag, possibly with a pure time delay. The PID block sets a CV
output to the process and uses the process feedback PV to determine an Error to adjust the
next CV output. A key process parameter is the total time constant, which is how fast does
the PV respond when the CV is changed. As discussed in the Setting Loop Gains section
below, the total time constant, Tp+Tc, for a first order system is the time required for PV to
reach 63% of its final value when CV is stepped. The PID block will not be able to control a
process unless its Sample Period is well under half the total time constant. Larger Sample
Periods will make it unstable.

The Sample Period should be no bigger than the total time constant divided by 10 (or down
to 5, worst case). For example, if PV seems to reach about 2/3 of its final value in 2 seconds,
the Sample Period should be less than 0.2 seconds, or 0.4 seconds worst case. On the other
hand, the Sample Period should not be too small, such as less than the total time constant
divided by 1000, or the Ki * Error * dt term for the PID integrator will round down to 0. For
example, a very slow process that takes 10 hours or 36000 seconds to reach the 63% level
should have a Sample Period of 40 seconds or longer.

Unless the process is very fast, it is not usually necessary to use a Sample Period of 0 to solve
the PID algorithm every PID sweep. If many PID loops are used with a Sample Period greater
than the sweep time, there may be wide variations in PLC sweep time if many loops end up
solving the algorithm at the same time. The simple solution is to sequence a one or more 1
bits through an array of bits set to 0 that is being used to enable power flow to individual PID
blocks.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

571

26.5

Determining the Process Characteristics

The PID loop gains, Kp, Ki and Kd, are determined by the characteristics of the process being
controlled. Two key questions when setting up a PID loop are:

How big is the change in PV when we change CV by a fixed amount, or what is the
open loop gain?

How fast does the system respond, or how quickly does PV change after the CV
output is stepped?

Many processes can be approximated by a process gain, first or second order lag and a pure
time delay. In the frequency domain, the transfer function for a first order lag system with a
pure time delay is:

PV(s)/CV(s) = G(s) = K * e **(-Tp s)/(1 + Tc s)

Plotting a step response at time t0 in the time domain provides an open loop unit reaction
curve:

Figure 380

The following process model parameters can be determined from the PV unit reaction
curve:

Process open loop gain = final change in PV/change in CV at time t0
(Note no subscript on K)

Process or pipeline time delay or dead time after t0 before the process output
PV starts moving

First order Process time constant, time required after Tp for PV to reach 63.2%
of the final PV

Usually the quickest way to measure these parameters is by putting the PID block in Manual
mode and making a small step in CV output, by changing the Manual Command %Ref+13,
and plotting the PV response over time. For slow processes, this can be done manually, but
for faster processes a chart recorder or computer graphic data logging package will help.
The CV step size should be large enough to cause an observable change in PV, but not so
large that it disrupts the process being measured. A good size may be from 2 to 10% of the
difference between the CV Upper and CV Lower Clamp values.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

572

26.6

Setting Parameters Including Tuning Loop
Gains

As all PID parameters are totally dependent on the process being controlled, there are no
predetermined values that will work; however, it is usually simple to find acceptable loop
gain.

Set all the User Parameters to 0, then set the CV Upper and CV Lower Clamps to the
highest and lowest CV expected. Set the Sample Period to the estimated process
time constant (above)/10 to 100.

Put block in Manual mode and set Manual Command (%Ref+13) at different values
to check if CV can be moved to Upper and Lower Clamp. Record PV value at some
CV point and load it into SP.

Set a small gain, such as 100 * Maximum CV/Maximum PV, into Kp and turn off
Manual mode. Step SP by 2 to 10% of the Maximum PV range and observe PV
response. Increase Kp if PV step response is too slow or reduce Kp if PV overshoots
and oscillates without reaching a steady value.

Once a Kp is found, start increasing Ki to get overshooting that dampens out to a
steady value in 2 to 3 cycles. This may require reducing Kp. Also try different step
sizes and CV operating points.

After suitable Kp and Ki gains are found, try adding Kd to get quicker responses to
input changes providing it doesn't cause oscillations. Kd is often not needed and will
not work with noisy PV.

Check gains over different SP operating points and add Deadband and Minimum
Slew Time if needed. Some Reverse Acting processes may need setting Config Word
Error Sign or Polarity bits.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

573

26.6.1

Setting Loop Gains Using the Ziegler and Nichols Tuning
Approach

Once the three process model parameters, K, Tp and Tc, are determined, they can be used
to estimate initial PID loop gains. The following approach provides good response to system
disturbances with gains producing an amplitude ratio of 1/4. The amplitude ratio is the ratio
of the second peak over the first peak in the closed loop response.

Calculate the Reaction rate:

R = K/Tc

For Proportional control only, calculate Kp as:

Kp = 1/(R * Tp) = Tc/(K * Tp)

For Proportional and Integral control, use:

Kp = 0.9/(R * Tp) = 0.9 * Tc/(K * Tp)

Ki = 0.3 * Kp/Tp

For Proportional, Integral and Derivative control, use:

Kp = G/(R * Tp) where G is from 1.2 to 2.0

Ki = 0.5 * Kp/Tp

Kd = 0.5 * Kp * Tp

Check that the Sample Period is in the range

(Tp + Tc)/10 to (Tp + Tc)/1000

26.6.2

The Ideal Tuning Method

The “Ideal Tuning" procedure provides the

best response to SP changes, delayed only by the

Tp process delay or dead time.

Kp = 2 * Tc/(3 * K * Tp)

Ki = Tc

Kd = Ki/4

if Derivative term is used

Once initial gains are determined, convert them to integers. Calculate the Process gain K as
a change in input PV Counts divided by the output step change in CV Counts and not in
process PV or CV engineering units. Specify all times in seconds. Once Kp, Ki and Kd are
determined, Kp and Kd can be multiplied by 100 and entered as integer while Ki can be
multiplied by 1000 and entered into the User Parameter %RefArray.

User Manual

Chapter 26

GFK-1645M

Dec 2019

PID Function

574

26.7

Sample PID Call

The following PID example has a sample period of 100Ms, a Kp gain of .4.00 and a Ki gain of
1.500. The set point is stored in %R0001, the control variable output in %AQ0002, and the
process variable is returned in %AI0003. CV Upper and CV Lower Clamps must be set, in this
case to 20000 and 4000, and an optional small Deadband of +5 and -5 has been included.
The 40-word RefArray starts in %R0100. Normally User Parameters are set in the RefArray,
but %M0006 can be set to reinitialize the 14 words starting at %R0102 (%Ref+2) from
constants stored in logic (a useful technique).

Figure 381

The block can be switched to Manual mode with %M1 so that the Manual Command, %R113,
can be adjusted. Bits %M4 or %M5 can be used to increase or decrease %R113 and the PID
CV and integrator by 1 every 100 ms solution. For faster manual operation, bits %M2 and
%M3 can be used to add or subtract the value in %R2 to/from %R113 every PLC sweep. The
%T1 output is on when the PID is OK.

User Manual

Chapter 27

GFK-1645M

Dec 2019

Reading and Writing Data in Flash Memory

575

Chapter 27:

Reading and Writing Data in
Flash Memory

For Release 3.0 and later VersaMax Micro PLCs, two Service Request commands can be used
to read or write flash memory during run time:

•

SVCREQ 52: Reads from Flash Memory into Reference Memory

•

SVCREQ 53: Writes from Reference Memory into Flash Memory

Chapter 22: gives details of using use these Service Request commands in the application
program.

This chapter describes the Logic-driven Write to Flash feature, and gives application
program examples.

27.1

Logic-Driven Write to Flash

For a VersaMax Micro PLC, one of the configuration choices is whether the CPU will read the
initial values from flash or RAM at power-up. If RAM is the choice, the CPU will read the
reference tables data from its RAM memory at power-up.

If “Read from flash” is the configured choice, the CPU will read

previously stored reference

tables from flash memory at power-up. In addition, Release 3.0 and later VersaMax Micro
PLCs (which includes all 20-, 40-, and 64-Point Micro PLCs) will also read from flash all data

that has been stored to flash using Service Request 53. This feature is called “Logic

-drive

write to flash”.

The logic-driven flash data area is cleared using the programmer. When storing an

application to the PLC, if “Write All To Flash” is configured, all data that has been written to

flash using Service Request 53 is automatically deleted. The data area can also be cleared by
using the flash-EEPROM tab and selecting Write All Data to Flash. This option is not available
during Run mode due to the time required to erase flash.

27.1.1

Data Quantities

For 20-, 40-, or 64-point Micro PLCs, the flash memory used for the Logic-driven write to
flash feature has a total size of 8192 bytes. Of this, 8 bytes are reserved and cannot be used.
So the maximum available capacity is 8184 bytes. For 14, 23, and 28-point Micro PLCs, the
Logic-driven user flash memory is 64kB (65536 bytes) and the maximum available capacity
is 65528 after 8 bytes are reserved.

Each execution of Service Request 52 or 53 can transfer:

•

1 to 10 words of %R, %AI, or %AQ reference data plus 6 bytes of command data. That
means a 1-word write to flash requires 8 bytes of flash memory, while a 10-word
write to flash requires 26 bytes of flash memory ((10 words X 2 bytes per word) + 6
bytes of command data).

User Manual

Chapter 27

GFK-1645M

Dec 2019

Reading and Writing Data in Flash Memory

576

•

1 to 10 bytes of %I, %Q, %M, %T, or %G reference data plus 6 bytes of command data.
A 1-byte write to flash requires 7 bytes of flash memory, while a 10-byte write to
flash occupies 16 bytes of flash memory.

Therefore, because each request requires 6 bytes of command data, the most efficient use
of flash is by transferring data in 10-word increments.

27.1.2

If Flash Memory Becomes Full

If Service Request 53 attempts to write more data to flash memory than the maximum
available, a fault is logged in the PLC Fault Table.

Fault Group

APPLICATION_FLT

Fault Action

DIAGNOSTIC

Error Code

USERFLASH_FULL

204

Error Message

Logic Driven Userflash Full

If a Logic Driven Userflash Full fault is logged in the PLC Fault Table, the PLC must be power
cycled.

After power-cycling the PLC, if another Logic Driven Userflash Full fault is logged, the
contents of flash memory must be completely erase

d by downloading new logic with ‘Write

All To Flash’ selected during download to PLC. After power

-cycling the PLC, data can again

be stored in flash until the Logic Driven Userflash Full

“fault

is logged again.

The same configuration and application program must be stored on RAM and PLC flash to

use this feature. It should not be done with configuration stored with ‘Read from flash’ set

and ladder program on RAM. If the program on RAM calls Service Request 53 to store
reference memory on flash, then the PLC is power-cycled, if the configuration says to read
data from flash, the data will be read but the application program may not be the same one
that called Service Request 53.

27.1.3

Important Note

Service request 53 (Logic driven write to flash) should be performed with caution after an
operation is performed that makes the application in RAM not equal to the application in
flash. The following are few such operations:

•

Word for word change or Run Mode store

•

Stop mode store of logic to RAM only

•

Logic and Registers stored to Flash during Stop mode store, but CPU is configured
to read Logic from RAM and Registers from Flash on power up.

Such operations may make the data stored by Service request 53 incompatible with the
application in flash and hence need to be used carefully.

User Manual

Chapter 27

GFK-1645M

Dec 2019

Reading and Writing Data in Flash Memory

577

Example 1: Writing Data to Flash Memory

This example writes 10 continuous bytes from PLC references %G0001 through %G0080 into
flash memory. In this example, the segment selector (56) is for byte access and so are the
offsets and lengths.

Figure 382

Example 2: Reading Data from Flash Memory

This example reads the same 10 continuous bytes from %G0001 through %G0080 from
flash into %G reference memory into the PLC.

Figure 383

EMERSON PACSystems VersaMax MICRO PLC & NANO PLC. USER MANUAL (2019) - page 38