MODBUS Client communication protocol

Supported device types and versions
Communication line configuration
Line protocol parameters
Station configuration
I/O tag configuration
Note to FloBoss 103 device
Note to Honeywell
Literature
Changes and modifications
Document revisions

Supported device types and versions

The protocol implements client (master) communication with arbitrary devices which support MODBUS RTU or MODBUS ASCII standards (serial communication) as well as MODBUS over TCP/IP. Moreover, it supports two extensions:

• Byte mode - allows working with devices that implement the registers as 1-byte variables (in contrast with Modbus standard in which the register value is 2 bytes).
• Variable mode - allows working with devices that implement the registers with different sizes than standard 2 bytes. It was implemented because of support of the flowmeter FloBoss 103 made by Fisher Controls International (at this time a part of Emerson Process Management): 1-byte variables, 4-byte unsigned/signed integers, text strings of length 10,12,20,40 characters, a 6-byte time stamp, and other.
• Passive (scanning) mode allowing to work in eavesdropping mode. This applies especially to serial communication when the communication port of the device is already used to communicate with another Master device. Due to the nature of the Modbus protocol, it is necessary to receive both requests and responses in this mode.

Communication line configuration

• Line category Serial (serial communication)
• Line category SerialOverUDP Device Redundant (serial communication).
• Line category RFC2217 Client (serial communication).
• Line category TCP/IP-TCP and TCP/IP-TCP Redundant (MODBUS over TCP/IP). Reserved TCP port 502 is commonly used, but it is possible to use any other one according to the setting of the device. The line number is not used, set the value e.g. to 1.
  Note: For redundant systems, it is possible to enter multiple names/addresses separated by commas.
  Note: In the case of WAGO 750-8100 type PLC and communication via MODBUS TCP, it was necessary to set a small polling period (e.g. 1 second) in the time parameters of the station. In the case of a longer period (5 seconds), the connection was closed quite often by the PLC.

Line protocol parameters

A dialog window of communication line configuration - Protocol parameters tab.
They influence some optional protocol parameters.

The line protocol contains the following parameters:

Station configuration

• Communication protocol "Modbus Client".
• The station address is a decimal number mostly in the range of 1 up to 247. Address 0 is reserved as broadcast.

Station protocol parameters

Configuration dialog box - tab Parameter.
They influence some optional parameters of the protocol. The following station protocol parameters can be set:

Table 1

I/O tag configuration

Possible types of I/O tag values for invariable mode: Ai, Ao, Di, Do, Ci, Co, TxtI.

Possible types of I/O tag values for variable mode: Ai, Ao, Di, Do, Ci, Cout, TxtI, TxtO, TiA.

I/O tag address:

The main address space in the protocol MODBUS is divided into the following registers:

• Coils type (reading/writing)
• Discrete Inputs (reading)
• Holding Registers (reading/writing)
• Input Registers (reading)

Independent addressing with the address size of 2 bytes, i.e. addresses from 0 up to 65535 (so-called MODBUS PDU addressing model), is in an address space of each type of register. Some devices work with address space starting with 1 (so-called MODBUS Data Model). In this case, it is necessary to deduct 1 in the address at configuration I/O tags in the D2000 system or change the setting of the parameter Addressing model to the MODBUS data Model.

The I/O tag with an address starting with %IGNORE will be ignored.

I/O tag address can be in a basic or extended format (for a variable mode).

Basic format of I/O tag address:
Address format is [I|U|Uu|Ul|f|F|L|Ll|S|Sl|B|X|sn.|an.|An.][d|D][b][s]RdFn[-WrFn[d]].Address[.BitNr] in which:

• The first character defines a type of I/O tag:
  • I - Integer16 (default) - one register is read, signed
  • U - Unsigned16 - one register is read, unsigned
  • Uu - Unsigned16 - one register is read, unsigned, only the upper byte is processed (1st in sequence)
  • Ul - Unsigned16 - one register is read, unsigned, only the lower byte is processed (2nd in sequence)
  • f - Float (4 bytes = 2 registers) - two registers with Address and Address+1 are read and transmitted as big-endian (see Note).
  • F - Float (4 bytes = 2 registers) - two registers with Address and Address+1 are read and transmitted as little-endian (so-called Modicon format), (see Note)
  • L - Unsigned long (4 bytes = 2 registers) - two registers with Address and Address+1 are read, unsigned, and transmitted as big-endian (see Note)
  • Ll - Unsigned long (4 bytes = 2 registers) - two registers with Address and Address+1 are read and transmitted as little-endian, unsigned (see Note)
  • S - Signed long (4 bytes = 2 registers) - two registers with Address and Address+1 are read, signed, and transmitted as big-endian (see Note)
  • Sl - Signed long (4 bytes = 2 registers) - two registers with Address and Address+1 are read and transmitted as little-endian, signed (see Note)
  • B - Byte unsigned, only the upper 8 bits of the register value
  • X - Byte unsigned, only the lower 8 bits of the register value
  • sn. - Text string with the length of n characters, one register is one character, n registers with Address up to Address+n-1 are read
  • an. - Text string with the length of 2*n characters, one register is two ASCII characters, characters are transmitted in the same order as they appear in the string,  n registers with Address up to Address+n-1 are read
  • An. - Text string with the length of 2*n characters, one register is two ASCII characters, characters are transmitted in big-endian order (i.e. "1234" is transmitted as "2143"),  n registers with Address up to Address+n-1 are read
• Modifier d indicates that a number is an 8-byte number (4 consecutive registers). It can be used for types L, Ll, S, Sl, F,f, and it is used for configuration of signed/unsigned 8-byte integer as well as an 8-byte float (big-endian <B8>..<B1> and little-endian <B1>..<B8> formats).
  Modifier D indicates that a number is an 8-byte number (4 consecutive registers). It can be used for types Ll, Sl, F and it is used for configuration of signed/unsigned 8-byte integer as well as an 8-byte float (little-endian format <B2><B1><B4><B3><B6><B5><B8><B7>).
  Note: when using a modifier d or D, the I/O tag must be of Analog type (Ai), because the Integer type (Ci) in D2000 is implemented as a 4-byte variable, and overflow might occur. Writing of Integer type (Co) as an 8-byte number is supported.
• Modifier b indicates that figure is coded by BCD. It can be used for I/O tags of I, U, B, L, Ll types.

• Modifier s indicates that a status register (Unsigned16) located on address Address is followed by a big-endian Float value located on address Address+1 .. Address+2. This indicator is used for type f and it is implemented for calorimeter Endress+Hauser RMS621. The following table shows the values of the status register and their mapping to D2000 attributes.

  Status register	D2000 attributes
  0: Invalid value	Weak
  1: Measured value valid	Valid
  2: Overflow warning 3: Overflow error 4: Underflow warning 5: Underflow error 6: Saturated steam alarm 7: Error in differential pressure calculation 8: Wrong medium for DP calculation 9: Wrong value range - DP calculation inaccurate 10: Differential pressure - general error 11: Range overshoot (Tsat > 350 etc.) on 12: Change in state of aggregation 26 : Differential pressure --> general error 99: No measured value is assigned to the register in the setup of the ModBus	Weak

• Parameter RdFn is a function of the Modbus protocol for a data reading. The following functions are implemented:
  • 1 - Read Coils: binary status reading
  • 2 - Read Discrete Inputs: binary input reading
  • 3 - Read Holding Registers: status register reading (Integer16/Unsigned16 and Float32 - reads two successive registers)
  • 4 - Read Input Registers: input register reading (Integer16/Unsigned16 and Float32 - reads two successive registers)
  • 0 - A value is not read, it is only written. The function for writing (WrFn) must be set.
• Parameter WrFn is the function of the Modbus protocol for data writing. The following functions are implemented:
  • 5 - Write Single Coil: binary status writing (default for Read Coils)
  • 6 - Write Single Register: status register writing (default for Read Holding Registers)
  • 16 - Write Multiple registers: multiple registers writing, it must be used when 2-register type is written (e.g. Float, Unsigned long, etc.).
    Note: function can be used to write more than two registers at once if a text string is used. Example:
    if we have an I/O tag with address a3.0-16.#8A00 (i.e. text string covering 3 registers, having length of 6 characters) and we write a string '123456', then hexadecimal values 0x3132, 0x3334 and 0x3536 (ASCII code for '1' is 0x31, for '2' is 0x32 etc) will be written to registers 0x8A00, 0x8A01 and 0x8A02.
  • 22 - Mask Write Register:  write affects only the value of the particular bit BitNr of the status register. It is usable only for Do value types with the address parameter BitNr.
• Parameter d activates the function "delayed write". Sending of the value is delayed until the request to write the value of the object without parameter d comes. All accumulated requests waiting to be written are sent. If the function WrFn is set to "Write Multiple Registers", the values are sent in one packet.
• Parameter Address is a 2-byte address of the register (0-65536). See also the protocol parameter Addressing model.
  Note: address can be specified as a hexadecimal number using a number sign (#), e.g. #50CE
• Parameter BitNr is a bit's position in a word. The values 0-7 are allowed to be used for binary statuses and inputs, values 0-15 are allowed to be used for reading of bit from 16-bit status or input registers.
  Note: coexistence of an I/O tag without BitNr parameter and multiple I/O tags with BitNr parameter having the same Address is possible.

Note about the byte and register order

1. MODBUS protocol uses the big-endian, i.e. the most significant byte (MSB) is transmitted as first. Examples:

2. When values are read from two registers as big-endian the received bytes are analyzed in this way:

Examples:

3. When values are read from two registers as little-endian, the received bytes are analyzed in this way (if Little Endian Mode=2143):

Examples:

Example of configuration:

• 1.10 - the function Read Coils reads the binary status value with address 10.
• 3.1 - a signed 16-bit number, it is read by the function Read Holding Registers from the address 1 (it can be also in the form I3.1).
• U3.1 - an unsigned 16-bit number that is read by the function Read Holding Registers from address 1.
• I3-6.1000 - signed 16-bit number that is read by the function Read Holding Registers from address 1000 and written by the function Write Single Register (as this function is the default, the address could be also I3.1000).
• S3.321 - a signed 32-bit number, it is read by the function Read Holding Registers from the registers 321 and 322.
• B1.20.0 - a bit that is read by function Read Coils from address 20 as 0-bit in a byte.
• s10.3.123 - a text string, length 10 characters (2 bytes per character), it is read by the function Read Holding Registers from the address 123.
• a5.3.123 - a text string, length 10 characters (1 byte per character), it is read by the function Read Holding Registers from the address 123.
• U0-6.456 - an unsigned 16-bit number, is written to the register 456, it is written by Write Single Register, a register reading is not performed.

Extended format of I/O tag address:

Address format is [xN].[I|U|F|B|C|T][b]RdFn[-WrFn].Address[.BitNr] in which:

• xN indicates the number of bytes that read or write. Valid values for N are 1, 2, 4, 8 (in combination with I, U, F), 6 for T type, and an arbitrary number for C type.
• A letter defines the type of I/O tag. Besides standard I, U, F, B, two extra types have been added:
  • C - text string of fixed length (e.g. x10.C3.1001 is a 10-character string on address 1001)
  • T - time stamp with length of 6 bytes (ss:mi:hh dd:mm:yy)
• The meaning of other parameters is in compliance with the standard mode.

See the example of the configuration in the next section.

Note to FloBoss 103 device

• configuration software ROCLINK800
• default login LOI, password 1000
• logging in FloBoss 103: click on DirectConnect (connection through COM1, on the side of FloBoss 103 it is connected to LOI-local interface)
• menu Configure->Modbus->Configuration
  set the parameter "Variable Mode" on station in D2000 according to setting "Byte Order":
  • if "Least Significant Byte first" then "Little endian"
  • if "Most Significant Byte first" then "Big endian"
• I/O tags are configured through menu Configure -> Modbus -> Registers on FloBoss 103
• following types are supported (n means 16-bit address):
  • Binary input:
    • address in D2000: 1.n, e.g. 1.1001, variable of Di/Dout type
    • address in FloBoss 103: variable of BIN type
      Function: 1
      Starting/ending register: n
  • Binary output:
    • address in D2000: 1.n, e.g. 1.1001, variable of Dout type
    • address in FloBoss 103: variable of BIN r/w
      Function: 1 (for reading)
      Starting/ending register: n
      Function: 5 (for reading)
      Starting/ending register: n
  • Unsigned Int 8 bits input:
    • address in D2000: x1.B3.n, e.g. x1.B3.1003, variable of Ci/Co type
    • address in FloBoss 103: variable of UINT8 type
      Function: 3A or 3B
      Starting/ending register: n
  • Unsigned Int 8 bits output:
    • address in D2000: x1.B3.n, e.g. x1.B3.1003, variable of Co type
    • address in FloBoss 103: variable of UINT8 r/w type
      Function: 3A or 3B
      Starting/ending register: n
      Function: 6
      Starting/ending register: n
  • Unsigned Int 16 bits input:
    • address in D2000: x2.U3.n, e.g. x2.U3.1004, variable of Ci/Co type
    • address in FloBoss 103: variable of UINT16 type
      Function: 3A or 3B
      Starting/ending register: n
  • Unsigned Int 16 bits output:
    • address in D2000: x2.U3.n, e.g. x2.U3.1004, variable of Co type
    • address in FloBoss 103: variable of UINT16 r/w type
      Function: 3A or 3B
      Starting/ending register: n
      Function: 6
      Starting/ending register: n
  • Signed Int 16 bits input:
    • address in D2000: x2.I3.n, e.g. x2.I3.1005, variable of Ci/Co type
    • address in FloBoss 103: variable INT16 type
      Function: 3A or 3B
      Starting/ending register: n
  • Signed Int 16 bits output:
    • address in D2000: x2.I3.n, e.g. x2.I3.1005, variable of Co type
    • address in FloBoss 103: variable of INT16 r/w type
      Function: 3A or 3B
      Starting/ending register: n
      Function: 6
      Starting/ending register: n
  • Unsigned Int 32 bits input:
    • address in D2000: x4.U3.n, e.g. x4.U3.1006, variable of Ci/Co type
    • address in FloBoss 103: variable of UINT32 type
      Function: 3A or 3B
      Starting/ending register: n
  • Unsigned Int 32 bits output:
    • address in D2000: x4.U3.n, e.g. x4.U3.1006, variable of Co type
    • address in FloBoss 103: variable of UINT32 r/w type
      Function: 3A or 3B
      Starting/ending register: n
      Function: 6
      Starting/ending register: n
  • Float 32 bits input:
    • address in D2000: x4.F3.n, e.g. x4.F3.1008, variable of Ai/Ao type
    • address in FloBoss 103: variable of FL type
      Function: 3A or 3B
      Starting/ending register: n
  • Float 32 bits output:
    • address in D2000: x4.F3.n, e.g. x4.F3.1008, variable of Co type
    • address in FloBoss 103: variable of FL r/w type
      Function: 3A or 3B
      Starting/ending register: n
      Function: 6
      Starting/ending register: n
  • String (N bytes) input:
    • address in D2000: x1N.C3.n, e.g. x10.C3.1010, variable of TxtI/TxtO type
    • address in FloBoss 103: variable of ACm(AC10,AC12,AC20,AC30,AC40) type
      Function: 3A or 3B
      Starting/ending register: n
  • String (N bytes) output:
    • address in D2000: xN.C3.n, e.g. x10.C3.1010, variable of Co type
    • address in FloBoss 103: variable of ACN r/w type (AC10,AC12,AC20,AC30,AC40)
      Function: 3A or 3B
      Starting/ending register: n
      Function: 6
      Starting/ending register: n
  • Time and date 6 bytes input:
    • address in D2000: x6.T3.n, e.g. x6.T3.1010, variable of TiA/TxtI type
    • address in FloBoss 103: variable of DT6 type
      Function: 3A or 3B
      Starting/ending register: n
    • Note 1: FloBoss 103 supports local and monotonous time - that is why the configuration of station in D2000 must correspond to configuration of FloBoss.
    • Note 2: It is possible to set time and date but it requires configuring extra I/O tags for a second, minute, hour, day, month, and year as Unsigned Int 8 bits and after that to write into them.

Note on Honeywell controllers

The basic parameters and current data of these controllers are not normally read by means of functions 0x01 up to 0x04. It is necessary to use the function 0x14/0x15 Read/write configuration reference data. These controllers use "big-endian" byte order. Therefore, for proper functionality, it is not necessary to modify parameters that change byte mode and endianness.

Examples of I/O tag configuration:

20.039 - 16-bit number from address 39(0x27)
f20.040 - 32-bit real number from address 40(0x28)

Literature

• MODBUS APPLICATION PROTOCOL SPECIFICATION V1.1b, December 28, 2006. http://www.modbus.org

Changes and modifications

-

Document revisions

• Ver. 1.0 - November 27th, 2006 - document creating.
• Ver. 1.1 - November 21st, 2007 - document update.
• Ver. 1.2 - April 24th, 2009 - document update.
• Ver. 1.3 - November 3rd, 2010 - document update.
• Ver. 1.4 - December 6th, 2010 - document update.
• Ver. 1.5 - September 5th, 2022 - document update (support for 8-byte values in the extended format).
• Ver. 1.6 - September 7th, 2022 - document update (added parameter "Dummy Request Mode").