Introduction

The RediGate Configuration Manual describes the configuration of many of the RediGate's standard features using the ACE program. This document gives additional instructions for configuring the RediGate to use the following protocol:

Rockwell Automation/Allen Bradley EtherNet/IP (EIP/CIP) Master

The EtherNet/IP master will communicate with ControlLogix, CompactLogix, and MicroLogix-820 PLCs. See the following Quick Start example configuration:

Ethernet/IP Master

To communicate with PLC5, SLC50x, or MicroLogix-1xxx PLCs, you will need to use the DF1 protocol instead.

EtherNet/IP Protocol Description

This document is not intended to provide a detailed description of the protocol(s) involved, nor to disclose proprietary information that may belong to other parties. Depending on the protocol, it may be necessary to refer to other vendor protocol documentation or device configuration details to understand how the RediGate should be configured to interface with it. This section provides a brief discussion of the protocol for the purpose of understanding the RediGate's configuration objects.

The Ethernet/IP protocol (EIP/CIP, an adaptation of Common Industrial Protocol, managed by ODVA) uses an object-oriented data model and employs both TCP/IP and UDP/IP as the transport protocols. An EtherNet/IP-compliant PLC or device will be configured with CIP objects having data that is typically defined in structures that may contain a mix of different data types and are typically identified by tag names. This differs from many other protocols supported by the RediGate, which commonly use a register-based data format. The RediGate's Poll Table is able to read CIP objects from the PLC and parse them into Real-time Database (RTDB) registers using the appropriate data types.

This document does not describe configuration of Allen Bradley DF1 or CSP protocols. The RediGate does not currently support Data Highway Plus (DH+), which requires a special hardware interface.

The following sections describe the ACE objects used for DF1 and CSP Master, object properties (including constraints on the Instance number), and object fields and their possible values required to configure for the given protocol(s). The object structure in ACE is hierarchical, with each object existing under a certain parent object. For instance, the FieldUnit is the child of one of several types of Circuit objects, either serial or network type (" System>Clients>Master Channels>Master Channel>Circuit").

The Description and Enabled properties are included in ACE as part of each object but are not mentioned here. The "UFF External" property is only mentioned for the objects where it is typically used, but should normally be left unchecked.

Supported Features and Limitations of EtherNet/IP Master

CIP objects may contain complex arrays and nested structures of data organization. In order to simplify the process of creating RediGate configurations, Elecsys provides a tool for converting the PLC configuration file (L5X) into the tables necessary for configuring the Scan Table, Poll Table, RTDB, and TagName tables.

The RediGate supports many standard features of EtherNet/IP, including:

RediGate can read the following Atomic data types: Bit, Int-8, Int-16, Int-32, Int-64, Real-32. A String is considered a structure.

RediGate can read the following Pre-defined Data Types (PDTs) of: TIMER, COUNTER, CONTROL, ALARM_ANALOG, ALARM_DIGITAL, PID, and STRING.

RediGate can read and write single data elements, arrays (single-dimensional only), UDT/PDT structures (up to 4000 bytes), array of Atomic elements within a structure, or a single structure within a structure (one level deep only).

RediGate currently cannot read PDTs that include ENCODED_DATA, MESSAGE, MODULE, ROUTINE, CONNECTION, PROGRAM, MOTION_GROUP, MOTION_INSTRUCTION, and AXIS.)

RediGate does not support structures within structures more than one level deep, and does not support an array of structures inside a structure.

Top-level arrays of structures will be read with one structure per poll. Groups of Atomic tags are scanned ten per transaction.
For example, a simple array of 10 timers, TimerArray[10], will execute ten separate polls of TimerArray[0] to TimerArray[9].
When reading large arrays, they will be broken into multiple blocks of 460 bytes.
Item tag names are imported from the L5X file. RediGate does not support automatic discovery of tags.
Data writes to Bit arrays are done by first reading 32 Boolean values from the RTDB, masking in the new values, and then writing all 32 Booleans to the PLC's 32-bit word.

Following are some additional limitations of the EtherNet/IP Master in the RediGate:

If elements are added or removed from a UDT structure, you should go through the L5X Import Tool to re-import the four tables into ACE (Scan Table, Poll Table, RTDB, TagNames, rather than modifying the tables by hand.
Implicit messaging (unsolicited data from device using RPI cycle time) via UDP is not supported.
Only peer-to-peer, single-level communications are allowed. No tunneling is allowed through PLCs to other PLCs, networks, or protocols using the EtherNet/IP master driver.
The EtherNet/IP Master is unable to identify "Local:1" physical I/O items. The PLC program must create tags or aliases for each I/O point required in the polling communication.

If an external Modbus host is reading PLC data from the RediGate using a Modbus Slave Channel, then there are special requirements for the organization of RTDB registers (see the "Modbus Slave Attach" section of the RediGate Configuration Manual).

Using the L5X Tool

See the Ethernet/IP Master Quick Start documentation for instructions on setting up a demo configuration to poll an EtherNet/IP device.

See Using the L5K Import Tool for brief instructions on using the ACE Configuration Management Utility to select PLC tags.

Or see Elecsys ACE Management Utility for more detailed instructions on using the ACM utility.

EtherNet/IP Master Channel

The structure of ACE objects for a Master Channel used for the EtherNet/IP protocol is shown below:

EtherNet/IP Master

RediGate Product Support > Protocol_EtherNetIP-Master > image2017-11-17 10:41:8.png

Note the following when configuring the Master Channel for EtherNet/IP devices:

Response Timeout needs to be set greater than the Scan Period - this is because the RediGate uses the Response Timeout to notify the EtherNet/IP PLC of what period it should expect to be polled (or else the PLC may shut down its open socket prematurely).
The Scan Table should be imported using the L5X Import tool in the ACM Utility. The PLC polling setup can get somewhat complicated, so it is not recommended to modify the table by hand.
If the configuration includes more than one EtherNet/IP device, either configure them under different Master Channels, or under separate Network Circuits.

Network Circuit

A Network Circuit is an IP network communications path to one or more field units from a common Master Channel. The Network Circuit is used when the field unit is connected to the RediGate over a TCP/IP network.

To configure the Network Circuit for an EtherNet/IP PLC:

Use Master Network Port = 44818.
Set the Circuit Type to be "Network Circuit."
Set the Connect Table to the IP address of the PLC.

See the RediGate Configuration Manual for information on configuring the NetCircuit object.

EtherNet/IP FieldUnit

An EtherNet/IP FieldUnit object contains unique information for each Field Unit using the EtherNet/IP communication protocol, and defines parameters for how data is read and written for the device.

If the PLC includes mixed data types, the RediGate uses a technique called "continuation polls" to optimize polling data. This method allows for one row in the Poll Table to be defined as the actual "poll" that requests data from the PLC, and subsequent rows (with Read Count=0) simply parse the data sequentially from the response obtained in the original poll.

The Continuation Poll may be used either to poll a structure, or it may be used to poll a block of Atomic elements that aren't in a structure. See the Poll Table description and Examples of EtherNet/IP Polls below for more details.

NOTE: The Scan Table in the Master Channel should only request the FIRST row in the Poll Table defining the "continuation poll" (with non-zero Read Count), not the subsequent rows of the continuation that have Read Count=0.

Attributes	Function
Object Type	FieldUnitEthernetIP
Parent(s)	System → Clients → Master Channels → Master Channel → NetCircuit
Instance	Must be unique under a Circuit.
The Field Unit must have an RTDB child object defined under it.

Properties	Values
Unit Name
Unit Address	Enter a unique field unit address under the Master Channel. For EtherNet/IP Master, the Unit Address is not actually used in the PLC. It is necessary in the RediGate to refer to the Channel/Unit, both in the Master Chanel Scan Table and in other processes that refer to the unit address.
Protocol	Placeholder for Protocol type of Ethernet-IP Master'.
Com Retries
Comm Status Holdreg
Produce RBEs
Poll Table
Controller Type	Select the controller type of the PLC being polled (ControlLogix, CompactLogix, MicroLogix 8xx).
Path Info	Select the slot number in the PLC for whether the CPU controller is installed, or "No Path" for a PLC without a slot (such as MicroLogix-820).
Max Read Msgs in-flight	Select the number of simultaneous in-flight messages to allow while polling. This only applies to the MicroLogix-8xx series of PLCs which don't support the Multi-read Service. MicroLogix-820 must be set to "1 in-flight message."
Vendor ID	Reserved for future use.
Spare	Reserved parameter

The Poll Table for EtherNet/IP should be imported using the L5X Import tool in the ACM Utility. The PLC polling setup can get somewhat complicated, so it is not recommended to modify the table by hand.

The '+' symbol may be used in the TagName column to look up an entry in the TagNames object (child of the RTDB for this FieldUnit) to associate an RTDB register with its configured tag name. Benefits of using the '+' are that the tag names only have to be defined in one place, and the TagNames object allows tags up to 100 characters.

Poll Table Properties
Properties	Values
TagName or Table-Ref	To poll discrete Atomic element(s) – Enter either the name of the Atomic Element (up to 40 characters), or enter '+' to look up the tag in the TagNames object. Multiple Atomic elements (not part of a Structure) may be polled in a block by using a Continuation Poll. To poll an Array – Tag name should be the name of the array, followed by the starting element in square brackets, such as: MyIntArray[3] It is not normally recommended to use the '+' for Arrays, because it's very inefficient (each array element would be polled as a separate tag). The starting element begins with 0, and the number of array elements to read should be specified in both the Read Count and Write Count (should be identical). An Array of Atomic elements may be included as the first row of a Continuation Poll that includes other Atomic elements, but it cannot be used as a secondary row in such a poll. An Array of Atomic elements may also be included as a row in a Continuation Poll of a Structure. An Array of UDTs is not allowed. To poll a Structure – Enter either the name of structure, or '+' to look up the Structure name in the TagNames object. Continuation Poll rows in a Structure poll would include the names of elements in the structure or '+' to look up in the TagNames object, such as: Structure.Element_0 Tag names with dots may be interpreted by some MQTT hosts as a nested tag name hierarchy. Tag names beginning with ten 'Z' characters are used in Allen Bradley PLCs to indicate hidden INT, SINT, or DINT values from which BIT fields are extracted.
Data Type	Select the data type of the data being read in this row of the Poll Table. If the Read Count of points contains data of different types, use multiple rows to define the number of points to read for each Data Type and the Destination Register in which to store each set of data. Select the data type from: Boolean (C1) – Use Boolean RTDB register type Sint8(C2), Sint16(C3), Uint8(C6), Uint16(C7) – Use UINT16 or SINT16 RTDB register type Sint32(C4), Uint32(C8) – Use UINT32 or SINT32 RTDB register type Sint64(C5), Uint64(C9) – Use SINT64 RTDB register type Real32(CA), Real64(CB) – Use REAL32 or REAL64 RTDB register type String(D0) - Use STRING32 or STRING256 RTDB register type BitArray(D3) - Read bit-packed integers. Use Boolean RTDB register type, and the Count should be the number of 32-bit entities, not the number of bits. *Struct(2A0)* - Read a UDT/PDT structure. The Read Count should be the total number of elements to read from the structure, and the Write Count should be set to 0 (use multiple rows following as Continuation Poll). LOG_COMMANDS_TO_STRING32 –
Read Count	For Atomic elements – Enter 1 for a single element or in the first row in a Continuation Poll. Enter 0 for the remaining Atomic elements in a Continuation Poll. It is not permitted to use a Read/Write count greater than 1 for reading individual Atomic elements – instead, use Continuation Poll rows to read multiple Atomic elements, one per row. For an Array – If polling a single Array, enter the number of array elements in both the Read Count and Write Count. If polling an Array as part of a Structure, enter 0 for Read Count on a Continuation Poll row, and enter the number of array elements in the Write Count. When polling an Array, no other individual Atomic elements can be part of its Continuation Poll. For a Structure – If polling for a structure, use multiple rows in the Poll Table as a "Continuation Poll." On the first row of a Continuation Poll, enter the Read Count as the size in bytes to request from the device (only a single Structure will be requested), and set the Write Count=0. This first Poll Record will not store any data into the RTDB. The following records of the Continuation Poll must have Read Count=0 and Write Count specifying the number of elements to store sequentially into the RTDB.
Write Count	For Atomic elements – The Write Count must be 1 when polling an individual element. For an Array – Enter the number of array elements to store into the RTDB. For a Structure – If polling for a structure, use multiple rows in the Poll Table as a "Continuation Poll." On the first row of a Continuation Poll, enter 0 for the Write Count. The first Poll Record will not store any data into the RTDB. The following records of the Continuation Poll should specify the number of Data Type elements to parse from the message and store sequentially into the RTDB beginning at the RTDB Destination Address. Typically, this will be 1 for an Atomic element or a count of elements for an Array. If parsing data using the "Bit" type, the count must be the number of binary bits. Include as many rows as needed to parse data from the poll message/structure, with Read Count=0 and Write Count set to parse the data type.
RTDB Destination Address	Enter the starting destination register within the field unit's RTDB into which the Write Count of data points will be stored. These registers must be configured in the RTDB object. The Destination Register type should be chosen based on the Data Type of the data points defined on this row of the Poll Table.
Skip Reg	The Skip Reg and following parameters allow an advanced level of control over the poll messages sent to the device. This feature allows logic (ISaGRAF, POD) or other controls to modify the polling of individual commands to the device. Enter an integer RTDB register address from this FieldUnit's RTDB which will be checked before issuing the poll record. The value in the register will be compared with the Skip Test and Skip Match properties. If the condition is satisfied, then the poll is handled according to the Skip Test setting. Enter 0 to disable this feature, or when using one of the last three Skip Test options (below).
Skip Test	Select a condition to use for poll modification. If a "Skip" condition is satisfied, then the Poll Record that would have otherwise been sent to the PLC is skipped, and the polling process moves immediately on to the next entry in the Master Channel's Scan Table. Equal Constant Match – If the value in Skip Reg register is equal to the Skip Match value (constant), the poll is skipped. If Grtr to Constant Match – If Skip Reg is greater than Skip Match (constant), the poll is skipped. If Less to Constant Match – If Skip Reg is less than Skip Match (constant), the poll is skipped. If Not Equal Constant Match – If the Skip Reg is not equal to the Skip Match (constant), the poll is skipped. If Equal RtdbReg Match – If the value in Skip Reg register is equal to the value contained in the register pointed to by Skip Match, the poll is skipped. If Grtr to RtdbReg Match – If the value in Skip Reg is greater than the value in the Skip Match register, the poll is skipped. If Less to RtdbReg Match – If the value in Skip Reg is less than the value in the Skip Match register, the poll is skipped. If Not Equal RtdbReg Match – If the value in Skip Reg is not equal to the value in the Skip Match register, the poll is skipped. NO SAVING Data with this Record – If using this Skip Test condition, the data elements in the Poll Table row are discarded and not stored in the RTDB. (Use 0 for Skip Reg and Skip Match.) This option can be used when there are points in a structure that don't need to be stored, so the protocol can par SAVE Integer to Hidden Buffer for Bit Extraction – When polling bit-packed integers, use this Skip Test condition to store the integer value into a hidden non-RTDB buffer temporarily for later extraction of bits. (Use 0 for Skip Test EXTRACT BIT from Hidden Integer Buffer – Must follow the "Save Integer to Hidden Buffer for Bit Extraction row in the Poll Table." These can include "ZZZZZZZZZZ" entries to discard unneeded bits. Read=Write=1.
Skip Match	For "Constant Match" Skip Test selections, enter a constant value that will be used for the match criteria. For "RtdbReg Match" Skip Test selections, enter the address of a register in this FieldUnit's RTDB which contains the value used for the match criteria.
Save Reply	Enter the address of a register in this FieldUnit's RTDB which will contain the length of the reply message (in bytes) from the PLC. This is only valid for the first row of a Continuation Poll. The Save Reply feature allows the length of a successful poll to be saved to the RTDB, which could be the same as a Skip Reg register used to skip a poll on the next poll cycle once it has been received successfully. Set the Save Reply column to 0 if you don't wish to store the length of the reply message.
Comment	The Comment column is an optional field, allowing a description to be entered for each row in the table. This is only for the user and has no effect on the operation of the RediGate.

Examples of EtherNet/IP Polls

The following examples show different cases of using the EtherNet/IP Poll Table.

Example 1: Continuation Poll of Multiple Atomic Values, plus an Array poll

The screen captures below show a Poll Table and TagNames table with multiple Atomic elements.

The Master Channel's Scan Table would only scan Row 1 of the table (containing Read Count=1). The RediGate would look up the RTDB addresses for the subsequent 4 rows with Read Count=0 and send out a poll for all the tags Temperature1, Pressure1, Pressure2, FlowRate, and FlowTotal. The resulting data would be stored in registers 40001, 40002, 40003, 47001, and 45001, respectively. This could also have been configured by putting the tag names directly into the 1st column of the Poll Table instead of the '+' symbol, but if MQTT is being used, the tags have to be entered in the TagNames table anyway.

RediGate Product Support > Protocol_EtherNetIP-Master > image2017-11-21 15:56:21.png

The above example also shows polling a single-dimensional array of Boolean alarms (not a Continuation Poll) and storing the first 4 values into registers 10001-10004. The Master Channel would need to scan Poll Record 6 to request this array.

Example 2: Read Structure using Continuation Poll

The screen captures below show an example Poll Table and TagNames table for polling a TIMER structure.

The first row includes the tag "TIMER_1", the name of the structure. This is the Poll Record scanned in the Master Channel.

Rows 2 and 3 handle hidden registers (Booleans packed into integers).

Row 4 parses and discards five bits from Timer4, then stores three Booleans from the timer (DN, TT, and EN) into 10001-10003, and two integers (PRE, ACC) into 45002-3.

RediGate Product Support > Protocol_EtherNetIP-Master > image2017-11-21 17:8:53.png

RTDB

See the RediGate Configuration Manual for information on configuring the RTDB.

The RTDB table should be imported using the L5X Import tool in the ACM Utility. The PLC polling setup can get somewhat complicated, so it is not recommended to modify the table by hand.

Tag Names

See the RediGate Configuration Manual for information on configuring the TagNames object.

The TagNames table should be imported using the L5X Import tool in the ACM Utility. The PLC polling setup can get somewhat complicated, so it is not recommended to modify the table by hand.

See Elecsys ACE Management Utility for more information on using the ACM Utility.

Table of Contents