BB4-8422 Modbus
TCP Server
Modbus TCP
Server Error Counts
The error counts illustrated on this page refer to
errors reported back to remote clients or masters about their attempt to
read/write registers in this IoTServer device.
Modbus TCP
Server Address Table
The server address table is used to define the set
of Modbus registers that will be visible to other Modbus clients (masters) when
the IoTServer is acting as a Modbus server (slave).
The set of Modbus registers that exist in the IoTServer
can be defined only once, but you have the option of remapping them to
different register numbers that will be accessed by secondary ports on the
device. Most often you would use the same register numbers whether they are
accessed from Modbus TCP or Modbus RTU, but it is possible to use a completely
different set of numbers for each different network port.
The address table can only be defined by the Primary
Server. All other instances of Modbus servers or slaves are secondary, and will
access the same set of Modbus registers with the option of remapping them to
different numbers.
The address table as it will appear for the Primary
Server is illustrated below.
The address table maps local objects to Modbus
registers. Note that in the above example, the Modbus registers skip one or
more numbers in some instances. Modbus holding registers are by definition
strictly a 16-bit entity. When a data element occupies more than 16 bits, it
requires multiple Modbus registers. Therefore a 32-bit value occupies 2
registers, while a 20-character string will occupy 10 registers (character
strings are packed 2 characters per register). The number of Modbus registers
assigned to each entry in the table is displayed in brackets in the format
column.
Modbus addresses generally refer to holding
registers. However, the server will provide access to the same register numbers
as input registers (for reading only), or as coils or discrete inputs (for
reading only). If a single bit register type such as coil is used to access a
register number, the value will be zero if the local object contains a zero,
and will provide a one if the local object contains anything other than zero.
The address table as it will appear for all
Secondary Servers is illustrated below. Note that the editing icons and buttons
are not available in the secondary address table.
Modbus TCP
Server Address Table Edit
There are two forms of the Modbus server address
edit page. When you click on the pencil icon on the address table page to
change a map, you will arrive at the following version of the address table
edit page:
If you click on the Add Maps button at the bottom of
the address table page, you will arrive at this version of the address table
edit page:
The parameters for either editing or adding Modbus
registers to the address table are as follows.
The following line will indicate "number",
"address", or "Modicon register",
depending on your preference which you set under User Settings.
Register Number or Address – Enter the number (starting at 1) or raw address (starting at 0) as
applicable. Do NOT enter 40001 for holding register 1 if you have not selected Modicon as the display format in your User Settings.
Modicon Register – Enter numbers like 40001 for the first holding
register if you have selected Modicon representation
in your User Settings.
The options available on the next line will vary
depending on selections made. The following are a few examples.
Register Format – Select the format of the data contained in the Modbus register(s), not
used by the protocol, but used by the gateway to interpret what the raw
increments of 1 or 16 bits should mean. Select format from the following table.
|
Format Label |
Format description |
|
“None” |
No format defined |
|
“Bit” |
Single bit, used
ONLY for Register Type Coil or Disc |
|
“Int” |
Integer (size and
whether signed are defined by labels below) |
|
“Real” |
Floating point
(single or double precision) |
|
“Char” |
Character string
with 2 ASCII characters per register |
|
“Mod10” |
Mod10 format, can
be 2, 3, or 4-register, specific to Schneider Electric meters |
Register Size – Register size refers to the number of consecutive input or holding
registers that should be read for a value greater than 16 bits. A 16-bit value
would have size of 1, a 32-bit value would have size of 2, and a 64-bit value
would have size of 4. Single precision Real (32-bit IEEE 754 floating point)
would be size 2, and double precision Real (64-bit IEEE 754 floating point)
would be size 4. If format is Mod10, then valid sizes are 2, 3, or 4 – check
manufacturer’s documentation if Mod10 is noted. Register “size” for a character
string will be character count divided by 2 (plus 1 of string length is an odd
number). Register Size is not used for Coil or Disc types.
IMPORTANT: If the register size is more than 1, then
the next assigned register address must be incremented by this count.
Unsigned – Select signed
or unsigned. Defaults to signed integer. Has no effect on Register Format other
than Int.
Endian Selection – Used when Register Size is greater than 1 to indicate what order the
registers should be interpreted in. Select "low" to indicate that the
lowest numbered register contains the least significant portion of data. Select
"high" to indicate that the lowest numbered register contains the
most significant portion of data. Although Modbus protocol itself is not
inherently “Little Endian”, many devices operate that way due to Intel
processors being inherently Little Endian. Modbus protocol does not stipulate
what the register order should be when multiple registers are treated as a
single data entity. Therefore, the user is required to pay attention to this.
Local Object – Specifies the local object number that contains the data that will be
provided to external Modbus clients or masters requesting this Modbus
"register".
The Rebuild button is a special case button for
rebuilding the entire address map. This is the easiest way to create a default
address table. Only the format information is used in this case, for example:
When using the Rebuild button, the starting number
will always be Modbus register 1, and the starting local object will always be
object 1. The rebuild will automatically assign the correct number of Modbus
registers, as determined by the format, to each local object. Modbus register
numbers will be assigned to all available local objects.
Modbus TCP
Server Remap Table
This page is where you have the option of remapping
the Modbus addresses or register numbers that some other Modbus client or
master would see when polling this device. Start by reviewing the Server
Address Table page if you haven't already.
Server Remapping Not Enabled
Note: If the server remapping option is not enabled
for this server in its Task Configuration, then the remap table will be empty and the Add Maps and Apply buttons
will be replaced with a message saying "Server Remapping Not
Enabled".
Modbus TCP
Server Remap Table Edit
The edit page for server address map is very simple.
Map local will indicate
"number", "address", or "Modicon
register", depending on your preference which you set under User Settings.
Register Number or Address – Enter the number
(starting at 1) or raw address (starting at 0) as applicable. Do NOT enter
40001 for holding register 1 if you have not selected Modicon
as the display format in your User Settings.
Modicon Register – Enter numbers like 40001 for the first holding register if you
have selected Modicon representation in your User
Settings.
As remote – The remote
Modbus register address or number that this local address should be viewed as
by remote clients or masters. Follow the same rules for number, address, or Modicon register as for the local value.
The Add Maps version of the remap edit page works as
follows. Let's assume you start with the empty table illustrated below. You
click the Add Maps button. 
The add/edit page is very simple. Follow the same
guidelines for local register numbers as noted above for editing. On this
version of the edit, you provide an offset instead of specific register number.
Each remapped number will be the local number plus offset. The other difference
on the Add page is that you also have an ending local number. Click Add to add
new entries for the entire range of addresses.
The result in our example is:
Modbus TCP
Server Port Settings
The ports page is only informative for Modbus TCP.
There is nothing to change here. You would change IP address of the IoTServer on the Network page. You would change task
configuration settings starting at the Task Configuration page.
Modbus TCP
Server Config File
All of your configuration information is stored in
an internal database each time you click the Save button on any page where
configuration entries may be made. To make configuration portable from one
device to another, and for purposes of retaining a backup copy, the
configuration information may be exported and imported as XML or CSV files.
This page is where your configuration file management takes place.
It is important to note that the XML file saved
within any one client/server function will contain the configuration
information for only that function. Depending on overall system configuration,
a complete backup may involve more than one XML or CSV file.
XML Files: When an XML file
has been selected, click the Load button to clear the configuration database
and reload configuration from the given XML file.
Select an existing name to overwrite or enter a new
file name, and then click Save to write the current configuration to the file
in XML format.
You may type in a new name in the file name window
for purposes of saving a new file. If you click the Refresh button, the file
name will be restored to the name currently loaded into the client. The name
could have been changed by selecting a file from the list below, or by typing
in a new name. If the displayed name has not yet been used, then Refresh will
restore the file name to what was most recently loaded.
CSV Files: If a CSV file is
selected, the Load and Save buttons will change into load/save CSV buttons.
When a CSV file has been selected, click the Load button to clear the
configuration database and reload configuration from the given CSV file.
Select an existing name to overwrite or enter a new
file name, and then click Save to write the current configuration to the file
in CSV format.
The drop-down list will show a list of all
configuration files currently found in the device's configuration folder. When
you select an XML or CSV file from this list, the name will be copied to the
Load/Save section of this page for pending load or save.
You may view the selected file by simply clicking
View. You can delete the file by clicking Delete.
You may upload files to the IoTServer
from your PC. Start by clicking Browse, and then use the browser's file dialog
to locate the file on your PC. Once a file is selected on your PC, click the
"Start upload" button to initiate the transfer.
You may also download files from the IoTServer to your PC. Click the Download button to transfer
the selected file to your PC.
Any time an XML or CSV file is loaded, an error log
file is generated. The error log file will be given the same name as the loaded
file, but with ".err" as the suffix instead of ".xml" or
".csv". You may view the error log by selecting it from the list and
clicking View.
Status is normally displayed in a message box at the
top of the screen when the load or save operation is complete. But if you want
to double check the status of the previous file operation, click Check Status.
The task (client or server) needs to be suspended
while a file load operation is in progress to prevent acting on any partial
configurations. This suspend/resume operation will normally happen
automatically as part of the sequence invoked by the Load button when loading
an XML file. The task must be explicitly suspended here for importing a CSV
file. The Suspend button will become a Resume button when the task is
suspended. Click Resume to continue operation. The current status is always
displayed here.
Example XML File
<?xml version="1.0"
encoding="ISO-8859-1"?>
<configuration>
<server_regs>
<reg addr="0" format="bit"
size="1" lowfirst="1" objnum="1"/>
<reg addr="1" format="int"
size="1" lowfirst="1" objnum="2"/>
<reg addr="2" format="int"
size="2" lowfirst="1" objnum="3"/>
<reg addr="4" format="int"
size="4" lowfirst="1" objnum="4"/>
<reg addr="8" format="int"
size="1" unsigned="1" lowfirst="1"
objnum="5"/>
<reg addr="9" format="int"
size="2" unsigned="1" lowfirst="1"
objnum="6"/>
<reg addr="11" format="int"
size="4" unsigned="1" lowfirst="1"
objnum="7"/>
<reg addr="15" format="real"
size="2" lowfirst="1" objnum="8"/>
<reg addr="17" format="real"
size="4" lowfirst="1" objnum="9"/>
</server_regs>
<remap_regs>
<map localAddr="0" remoteAddr="100"/>
<map localAddr="1" remoteAddr="101"/>
<map localAddr="2" remoteAddr="102"/>
<map localAddr="4" remoteAddr="104"/>
<map localAddr="8" remoteAddr="108"/>
<map localAddr="9" remoteAddr="109"/>
<map localAddr="11" remoteAddr="111"/>
<map localAddr="15" remoteAddr="115"/>
<map localAddr="17" remoteAddr="117"/>
</remap_regs>
</configuration>
Each line in the server_regs portion of the XML file defines one locally
known Modbus “register” which may actually span multiple Modbus addresses.
NOTE: Register configuration does
not specify a Modbus register type. This is because all register addresses in
the IoT Server can be accessed as any of the standard register types (coil,
discrete input, input register, or holding register). Therefore, the value at
input register address 0 and holding register address 0 will be the same value.
When referenced as a coil or discrete input the same address will appear as 0
if the value is 0, or 1 if the value in anything non-zero. All local registers
are 16 bits regardless of how accessed. The 16-bit value will be processed as 0
or 1 for coil or discrete input. Therefore, writing a 1 to an address as a coil
will result in a 16-bit integer 1 when the same address is accessed as a
holding register.
IMPORTANT: There can be only one
definition of a primary register map, but there can be multiple instances of a
Modbus server. The secondary servers can only remap the registers defined by
the primary server. The primary definition is found in the <server_regs> section while remapping is found in the
<remap_regs> section. A secondary server will
disregard the <server_regs> section. The
primary server can be either a Modbus RTU Slave or Modbus TCP Server.
XML attributes for each
register:
addr=”n” – (Json “address”, integer) – Modbus address in
the range of 0..65535. Note that Modbus address 0 is most often referred to as
“Register 1” or Modicon 40001 if it is a holding
register.
format=”xxx” (Json “format”, character string, and “formatCode”, integer) – Specifies the Modbus format in
which data should be interpreted. Valid formats for XML and Json are shown
below:
|
XML value |
Format
description |
|
“none” |
No format defined |
|
“bit” |
Single bit (coil, discrete only) |
|
“int” |
Integer (16-, 32-, or 64-bit) |
|
“real” |
Floating point (single or double) |
|
“char” |
ASCII character string |
|
“mod10” |
Schneider Electric Mod10 format |
size="n" – (Json “size”, integer) – Specifies the
number of 16-bit address locations that make up this “register”. Valid sizes by
register format are as follows:
|
Type |
Number of
registers |
|
Bit |
1 |
|
Integer |
1, 2, 4 (for 16-, 32-, 64-bit) |
|
Real |
2, 4 (for single, double precision) |
|
Character |
1..63 (registers - 2 characters per register) |
|
Mod10 |
2, 3, 4 |
Note: The maximum size for ‘char’
is 63 registers, which means 126 characters is the maximum string length. Size
is in registers, not characters.
unsigned=”n” – (Json “unsigned”, integer) – Applies
only to integer, and specifies that the integer should be treated as unsigned
in “n” is “1”. Integers default to signed (n=0).
lowfirst=”n” – (Json “littleEndian”,
integer) – Applies to any register greater than 16 bits (except “char”),
meaning the data value consists of 2 or more 16-bit registers. When lowfirst n=1, the least significant data will be found in
the lowest numbered or first Modbus register address of the “register” that
spans multiple addresses. When lowfirst n=0 (or
omitted), the most significant data will be found in the lowest or first Modbus
register address. For “char” registers, the start of the string will always be
in the lowest or first Modbus address.
objnum=”n” – (Json “localObject”,
integer) – Specifies the IoT Server global data object that will be provided
via this Modbus register. The global object data type and Modbus register data
format do not need to match. Data will be converted as applicable, including
numeric to character string conversion.
Each line contains simply a local
address and a remote address. The local address is expected to reference a
register previously created in the server_regs part
of the file (by the primary server). The remote address is any valid Modbus
address that will be used to reference this local register. There may fewer remap_regs entries than server_regs
entries, but generally not more.
localAddr=”n” – (Json “localAddr”,
integer) – Specifies local Modbus address, namely the Modbus address
established by the Register Configuration above.
remoteAddr=”n” – (Json “remoteAddr”,
integer) – Specifies the address of the same register as seen by a remote
client (master) querying our local Modbus device.
The following illustrates a CSV file for the primary instance of Modbus Server, which can be assigned to either an RTU or TCP channel. There can be only one instance of a primary server, and this instance establishes the correlation of local objects to Modbus registers. Secondary instances of Modbus servers will refer to the primary server’s mapping of objects to register addresses, optionally with remapping (see below).
Begin,Modbus,ServerMaps
LocalObj,RegAddr,RegFormat,RegSize,Unsigned,LittleEnd
1,0,INT,1,Y,N
2,1,INT,1,N,N
3,2,INT,4,N,Y
4,6,REAL,2,N,Y
5,8,REAL,2,N,Y
6,10,CHAR,10,N,N
7,20,CHAR,10,N,N
End
LocalObj – Specifies the local object number that contains the data that will be provided to external Modbus clients or masters requesting this Modbus “register”.
RegAddr – Raw 0-indexed address to be assigned to this register.
RegFormat – Format of the data contained in the Modbus register(s), not used by the protocol, but used by the gateway to interpret what the raw increments of 1 or 16 bits should mean. Select format from the following table.
|
Format Label |
Format
description |
|
“None” |
No format defined |
|
“Bit” |
Single bit, used ONLY for RegType Coil or Disc |
|
“Int” |
Integer (size and whether signed are defined by labels below) |
|
“Real” |
Floating point (single or double precision) |
|
“Char” |
Character string with 2 ASCII characters per register |
|
“Mod10” |
Mod10 format, can be 2, 3, or 4-register, specific to Schneider Electric meters |
RegSize – Register size refers to the number of consecutive input or holding registers should be read for a value greater than 16 bits. A 16-bit value would have size of 1, a 32-bit value would have size of 2, and a 64-bit value would have size of 4. Single precision Real (32-bit IEEE 754 floating point) would be size 2, and double precision Real (64-bit IEEE 754 floating point) would be size 4. If format is Mod10, then valid sizes are 2, 3, or 4 – check manufacturer’s documentation if Mod10 is noted. Register “size” for a character string will be character count divided by 2 (plus 1 of string length is an odd number). RegSize is not used for Coil or Disc types.
IMPORTANT: If the register size is more than 1, then the next assigned register address must be incremented by this count.
Unsigned – Indicate “Y” if unsigned, or “N” if signed. Defaults to signed integer. Has no effect on RegFormat other than Int.
LittleEnd – Used when RegSize is greater than 1 to indicate what order the registers should be interpreted in. Enter “Y” to indicate that the lowest numbered register contains the least significant portion of data. Enter “N” or omit to indicate that the lowest numbered register contains the most significant portion of data. Although Modbus protocol itself is not inherently “Little Endian”, many devices operate that way due to Intel processors being inherently Little Endian. Modbus protocol does not stipulate what the register order should be when multiple registers are treated as a single data entity. Therefore, the user is required to pay attention to this.
The following illustrates the CSV file or section for remapping register addresses in an instance of a Modbus server. The remapping can be optionally included in the primary instance, but is usually more commonly used in a secondary instance of server. The remapping effectively creates a different view of the same set of registers defined by the primary server. For example, in the CSV below, Modbus register 0 is seen at address 100 by any client requesting registers from this instance of the server. Different instances of the server will normally run on different ports in the system.
Begin,Modbus,ServerRemaps
LocalAddr,RemoteAddr
0,100
1,101
2,102
6,106
8,108
10,110
20,120
End
LocalAddr – The local Modbus register addressed assigned by the primary server. This address must exist in the register assignments made by the primary server.
RemoteAddr – The remote Modbus register address that this local address should be viewed as.