TM950

POWERLINK
configuration and
diagnostics
2 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Requirements
TM210 - Working with Automation Studio
Training modules TM213 - Automation Runtime

Automation Studio 4.11 or higher

Software

Hardware ETAL950 (POWERLINK) with X20CP1586

 TABLE OF CONTENTS 3

Table of contents
1 Introduction......................................................................................................................................................... 4
1.1 Learning objectives.............................................................................................................................. 5
1.2 Safety notices and symbols...............................................................................................................5
2 POWERLINK - Introduction...............................................................................................................................6
2.1 Industrial Ethernet............................................................................................................................... 6
2.2 POWERLINK........................................................................................................................................... 6
3 POWERLINK – Technology................................................................................................................................9
3.1 POWERLINK in the OSI model........................................................................................................... 9
3.2 Data link layer..................................................................................................................................... 10
3.3 CANopen application interface....................................................................................................... 15
3.4 Network management (NMT)......................................................................................................... 25
4 POWERLINK – Diagnostics.............................................................................................................................29
4.1 Diagnostics tools............................................................................................................................... 29
4.2 Startup behavior of a POWERLINK node...................................................................................... 39
4.3 Diagnostics strategy........................................................................................................................ 44
5 POWERLINK - Design and optimization......................................................................................................46
5.1 Network design and timing performance.................................................................................... 46
5.2 Automation Studio settings for network timing........................................................................ 53
5.3 Mapping in Automation Studio...................................................................................................... 56
5.4 Optimization options in Automation Studio............................................................................... 57
6 POWERLINK - Third-party device support..................................................................................................66
6.1 XDD - Device description file.......................................................................................................... 66
6.2 Device description file for modular devices................................................................................ 67
6.3 XDD application example: openPOWERLINK MN........................................................................68
6.4 Examples of non-B&R MNs:.............................................................................................................68
7 Summary............................................................................................................................................................ 69
8 Appendix – Solutions to the exercises........................................................................................................70
4 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

1 Introduction
When it comes to fieldbus technology, B&R relies heavily on the real-time, Ethernet-based POWERLINK network to
handle the highest demands on timing and data throughput. Automation Studio makes it easy to use this technology
in the field. In order to take advantage of all of the possibilities offered by the POWERLINK communication protocol
and ensure it is used correctly, it is important to learn as much as possible about the design, configuration and trou-
bleshooting options presented by this technology. This training module is designed to provide this information in the
form of extensive theory and practical examples that will offer a much deeper look into what comprises POWERLINK
technology.

This training module will explain the many different POWERLINK configuration and optimization options available in
Automation Studio and their effects on the overall system. It will also introduce a range of powerful tools for trou-
bleshooting errors on a POWERLINK network.
 INTRODUCTION 5

1.1 Learning objectives

This training module uses selected examples illustrating typical application tasks to help participants learn how POW-
ERLINK technology works and how it is applied in Automation Studio. It also presents the various options for diagnos-
ing errors and explains how to integrate 3rd-party devices.
Learning objectives and contents of this training module:
■ Participants will learn the basics of POWERLINK technology.
■ Participants will become familiar with the "CANopen" application interface for POWERLINK.
■ Participants will learn how to use service and diagnostics options.
■ Participants will learn how POWERLINK functions and is displayed in Automation Studio.
■ Participants will learn how to use the POWERLINK library in Automation Studio.
■ Participants will learn the meaning of the parameters for setting up networks.
■ Participants will learn about the possibilities for optimizing the POWERLINK network in Automation Studio.
■ Participants will learn how to integrate 3rd-party devices using a device description file (XDD).

1.2 Safety notices and symbols

The meaning and usage of safety notices and symbols is described in the main module "TM210 – Working with Au-
tomation Studio".
Special notes on this manual

Help: Unless otherwise mentioned, this note type always refers to Automation Help.
 6 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

2 POWERLINK - Introduction
POWERLINK is an Ethernet-based, real-time capable network protocol. It builds upon the physical and media access
control layers defined in the IEEE 802.3 standard and allows for deterministic transmission of payload data in the
microsecond range. Since POWERLINK builds upon the Ethernet standard without violating it in any way, it can be
implemented on any new or existing hardware/software platform.
POWERLINK is primarily used to transfer process data in automation systems.

2.1 Industrial Ethernet

Communication between the machines and components that make up an automation system takes place over fieldbus
networks. These are generally serial (RS485) or CAN-based systems with additional media access control and applica-
tion layers.
These types of conventional fieldbuses are, for example:
■ CANopen
■ PROFIBUS
■ DeviceNet

With more and more applications exceeding the performance capabilities of these serial and CAN-based systems,
the need for an alternative became increasingly urgent. With its widespread use in IT systems and 30-year history
of ongoing development, as well as interference-resistant cables ideal for harsh industrial environments, Ethernet
technology presented the ideal solution.
In the ISO OSI model1 (see 3.1 "POWERLINK in the OSI model"), the Ethernet standard describes layers 1 and 2, i.e.
the physical and media access control layers. The CSMA/CD2 method used here is ideal for a decentralized network
structure, but does not provide deterministic transmission. Deterministic transmission is an essential requirement
for industrial applications (particularly in control and positioning applications).

Figure 1: CSMA/CD mechanism

One way to prevent collisions on the network is using switches, which provide a separate bus segment for each sta-
tion. The disadvantage of this approach is that it increases the transmission delay of payload data. Especially short
throughput times in industrial applications, however, are more important than the actual data throughput.
It is also possible to use a request/response method. This approach defines a single network station that will manage
media access. In this scenario, all other stations are only permitted to send data after receiving permission.

2.2 POWERLINK

To solve the problem of deterministic transmission, POWERLINK uses a request/response, or "polling" approach. One
node3 – known as the master, or managing node (MN) – controls access to the network. The slaves, or controlled nodes
1 Open Systems Interconnection Model

2 Carrier Sense Multiple Access / Collision Detection

3 Network station addressed via a node number

 POWERLINK - INTRODUCTION 7

(CNs), follow the commands of the managing node and only transmit data upon request from the managing node.
This eliminates collisions4 and provides a basis for real-time capability.

Figure 2: Master-slave system

Standardization
POWERLINK is an open real-time Ethernet protocol, which means both the specification and an implementation –
known as a "protocol stack" or simply "stack" – are freely available. There are no licensing fees required. The licensing
model for the stack is a BSD5license, which provides the maximum amount of freedom. Development and maintenance
of the POWERLINK specification6 is handled by the Ethernet POWERLINK Standardization Group (EPSG). For more in-
formation, see www.ethernet-powerlink.org.
POWERLINK is both an IEC and IEEE standard as well as a Chinese national standard for communication technolo-
gy:
4 IEC 61784-2: Industrial communication networks - Profiles - Part 2: Additional fieldbus profiles for real-time net-
works based on ISO/IEC 8802-3.
4 IEC 61158: Industrial communication networks - Fieldbus specifications
4 IEEE 61158-2017: IEEE standard for industrial hard real-time communication
4 GB/T 27960-2011: Ethernet POWERLINK communication profile specification

4 Except in the event of an error (i.e. incorrect cabling or network settings)

5 Berkeley Software Distribution; open source license with no reciprocal obligation to disclose source code

6 www.ethernet-powerlink.org / Downloads / Technical Documents / EPSG DS 301 V1.x.x - Communication Profile Specification
8 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Characteristics of a POWERLINK network

• POWERLINK is based on Fast Ethernet (100 Mbit/s)
• Up to 239 CNs
• 200 µs7 minimum cycle time
• Recommended cable types:
■ Copper cable: Cat. 5, S/UTP, 26 AWG
■ Fiber optics: Multimode fiber with 62.5/125 µm or 50/125 µm core diameter

• Process data size per node: Up to 1,490 bytes8

• Asynchronous data transmission: Up to 1500 bytes per cycle and station9
• Topology: Line, star, tree or any combination thereof

7 Value achieved in established MN implementations (e.g. by B&R). The theoretical limit is lower, and is not defined in the specification.

8 A higher value can be reached with the extension "Multi PReq/PRes."

9 A higher value can be achieved with the"Multi ASnd" specification extension.

 POWERLINK – TECHNOLOGY 9

3 POWERLINK – Technology
3.1 POWERLINK in the OSI model

The OSI model is a standardized conceptual model of an open information processing system. The most significant
manufacturer-independent transfer protocols commonly used today are based on this model. It groups the internal
functions of a communication system into seven layers. Each layer builds upon the layer below it and represents an
instance that performs its task according to specific rules. These rules are defined by protocols, which may span mul-
tiple layers.
Layer 1, the physical layer, describes the electrical, mechanical and functional interfaces to the physical transmission
medium. The next layer above it, the data link layer, provides a reliable link between two directly connected nodes by
detecting errors that may occur in the physical layer. These two layers are often referred to collectively as "the physical
layers".
Layers 3 and 4 above them are known collectively as "the transmission layers". Layer 3, the network layer, controls the
logic and timing of data transmission. Layer 4, the transport layer, segments the data and allocates it to the applica-
tions.
Layers 5 (session layer), 6 (presentation layer) and 7 (application layer) are referred to collectively as "the application
layers" since all programs and applications access them directly. As explained above, Ethernet only defines the physical
and data link layers. Protocols that do not access the same layers can be used in combination – like POWERLINK and
CANopen.

POWERLINK specifies the data link layer and the layers above it, while CANopen deals exclusively with
the application layers.

Figure 3: POWERLINK application interface – CANopen

10 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

3.2 Data link layer

The POWERLINK data link layer (DLL) defines the mechanism for network access. Network nodes, so-called POWERLINK
nodes, are only permitted to send data at defined times. Network access is managed by a dedicated node, called the
managing node (MN). A POWERLINK network always has exactly one active managing node (MN). The MN can manage
up to 239 controlled nodes (CNs).

3.2.1 POWERLINK cycle

The POWERLINK cycle is divided into an isochronous and an asynchronous phase. In the isochronous phase, process
data from the network nodes is transferred cyclically. The asynchronous phase allows the exchange of acyclic data
between two or more nodes. This phase is used for network management, configuration, diagnostics and non-time-
critical data queries. It also allows for the transfer of any non-POWERLINK frames (e.g. IP-based data such as websites).

Figure 4: POWERLINK cycle

Abbrevia- Name Source Meaning

tion
SoC Start of cycle MN The MN sends this frame at the beginning of the POWERLINK cycle.
This synchronizes all of the network stations.

PReq Poll request MN The MN sends this frame along with the payload to each CN.

PRes Poll response CN/MN A CN sends a poll response with real-time data as a response to a
poll request.

SoA Start of asynchro- MN Marks the end of cyclic data communication and thus the start of
nous asynchronous communication and assigns transmission permis-
sions for asynchronous communication.

ASnd Asynchronous send CN/MN This frame transports asynchronous (non-real-time) data.

AMNI* Active Managing MN In networks with redundant masters, a MN uses this frame to indi-
Node Indication, cate that it is switching from passive to active mode to take over for
used by EPSG DS302- a failed MN.
A

Table 1: POWERLINK frame types

 POWERLINK – TECHNOLOGY 11

Abbrevia- Name Source Meaning

tion
AInv* Asynchronous Invite, MN This frame is used to assign additional transmission rights for asyn-
used by EPSG DS302- chronous communication and is used for the "Multiple Asynchro-
B nous Send" POWERLINK extension.

Table 1: POWERLINK frame types

*) These frame types are only included in the table for the sake of completeness. They are not shown in the image
above because they represent optional POWERLINK extensions.

The structure of a POWERLINK frame is shown in the image below (Fig. 5). The individual layers (from the OSI model)
encapsulate each other. A higher layer is always embedded in the payload data of the layer below it.
A POWERLINK frame is identified as such by value "0x88AB" in field "Ethertype" in the Ethernet frame header10 .

Figure 5: POWERLINK frame structure

3.2.2 Isochronous phase

In the isochronous phase the network stations exchange data cyclically. This is where real-time data traffic occurs.
Cycle time
The isochronous phase is initiated by the MN. The MN sends the Start of Cyclic (SoC) frame to all network stations.
The SoC is also used to synchronize the network. In a fully synchronized network, this frame activates the last received
output data and triggers sampling of input data. The SoC is sent as a multicast frame11. The cycle time is the time
between SoC frames sent by the MN.

10 "See Ethertype 0x0800 (identifies IPv4 frames)

11 Type of frame that is received simultaneously by an entire group of nodes (all POWERLINK nodes)
12 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

t Cycle t Cycle

MN SoC SoC SoC

Figure 6: The MN determines the cycle time by sending the SoC frame.

Direct cross-traffic
The poll request (PReq) and poll response (PRes) frames are also sent in the isochronous phase. PReq is a unicast
frame containing the MN's real-time data, generated only by the MN and sent to a specific CN. The respective CN must
respond to a received PReq with a PRes. The PRes contains the CN's real-time data. Since the PRes is defined as a
multicast frame, the CN's real-time data is distributed throughout the entire network. This makes it possible for a CN
to read data from another CN (in the PRes frame) directly from the network. Since this type of communication occurs
directly, without passing through the MN, it is called direct cross-traffic.

Figure 7: Direct cross-traffic with (POWERLINK) broadcast mechanism

Advantages of direct cross-traffic

■ Data transfer in one cycle, i.e. reduced response time
■ More cyclic nodes possible (without increasing cycle time)
■ Shortens the minimum required POWERLINK cycle time (with same number of nodes)
■ Reduces the load on the MN

Completing data transfer in one cycle makes it possible to achieve the lowest possible latency in CN-to-
CN communication.
One application of cross-traffic is for electronic axis synchronization.

Poll response chaining

Since the MN can also send a PRes, the same process data can be distributed to multiple CNs in a single step (broadcast
channel). This mechanism is required for poll response chaining mode. CNs operated in this mode no longer require a
PReq frame from the MN. Instead, they each respond to the PResMN frame with a constant delay. The delay for each
node is automatically calculated at startup12.
12 Special POWERLINK frames are used to calculate the timing for the topology (SoA with SyncReq flag, ASnd with SyncResp flag).
POWERLINK – TECHNOLOGY 13

Figure 8: Poll response chaining

Advantages of poll response chaining

■ More cyclic nodes possible (without increasing cycle time)
■ Shortens the minimum required POWERLINK cycle time (with same number of nodes)
■ Reduced topology-related latency (particularly with line topology)

The benefits of poll response chaining are primarily due to the elimination of one PReq frame per station.
This optimization is used in central closed-loop control systems where small volumes of data are trans-
ferred, e.g. in CNC systems (position control).

Multiplexing
To optimize the cycle time and better represent the real application in the network, nodes can also be operated in
multiplexing mode. This means that each CN is only addressed by the MN in every nth cycle.

Multiplexing prescale

Cycle i Cycle i+1 Cycle i+2 Cycle i+3

1 2 3 4 5 1 2 3 6 7 1 2 3 8 1 2 3 4 5

Standard: 1, 2, 3 Multiplexed: 4 - 8

Time

Figure 9: Multiplexing diagram

 14 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Advantages of multiplexing
■ Optimizes the use of bandwidth
■ More cyclic nodes possible (without increasing cycle time)
■ Shortens the minimum required POWERLINK cycle time (with same number of nodes)

Multiplexing can increase the maximum number of stations that are possible within one POWERLINK
cycle. Since multiplexing nodes can read PRes frames from other nodes at any time, a combination of
multiplexing and cross-communication can result in highly optimized solutions.
This optimization is used when transmitting status and configuration data (e.g. temperature measure-
ment) used in slow control algorithms.

A CN can be operated in "Standard" (PReq/PRes in every cycle), "Multiplexed" or "Poll Response Chaining"
mode. "Multiplexed Chaining" mode is not an option, i.e. it is not possible to set both modes at once.

3.2.3 Asynchronous phase

The asynchronous phase follows immediately after the isochronous phase (see Fig. 4). The MN initiates this phase by
sending the Start of Asynchronous (SoA) frame. Like the SoC, the SoA is also a multicast frame. In this frame, the MN
grants transfer permission to any station13 in the network (Fig. 10). The selected node may now send either asynchro-
nous POWERLINK data (ASnd) or an Ethernet frame of some sort (e.g. IP data). It is up to the node whether it is sent
as a unicast, multicast or broadcast frame.
The MN grants prioritized transfer permission based on previous access requests from network stations. These re-
quests to the MN can be contained in both asynchronous (ASnd) and cyclic (PRes) frames.
The asynchronous phase makes "hot plugging" possible, allowing CNs to be added to the network without affecting
its real-time behavior.

Figure 10: Assignment of an asynchronous slot

The main application of asynchronous communication is to transfer network management commands

(NMT) and asynchronous access to objects (SDO protocol).

Multiple asynchronous send

Multiple asynchronous frames can be sent in one cycle in order to increase throughput in the asynchronous phase.
Each CN must still be invited by the MN, but this must be done using a different type of frame – the Asynchronous
Invite (AInv). The MN does not invite itself with an AInv frame, but rather sends its asynchronous data directly in the
"extended" asynchronous phase.

13 i.e. even to itself

 POWERLINK – TECHNOLOGY 15

MN POWERLINK Cycle

Asynchronous phase

SoC PReq PReq SoA AInv ASnd ASnd AInv

PRes PRes ASnd ASnd ASnd

CN

Figure 11: Multiple asynchronous send

When the goal is to optimize the throughput of non-real-time data on a POWERLINK network, multiple
asynchronous send is used.
This optimization is used, for example, when transmitting image data that requires a large amount of
bandwidth.

3.3 CANopen application interface

-The POWERLINK application interface is based on CANopen14. Therefore, POWERLINK can also be referred to as
"CANopen over Ethernet". Although calling it this oversimplifies matters somewhat and fails to highlight POWERLINK's
real-time capabilities, it does capture the fundamental nature of the communication system.
This section describes the structure of the CANopen interface and explains how it is used to establish a data connec-
tion between POWERLINK nodes.

3.3.1 Object dictionary

The structure of the CANopen application interface uses an object dictionary (OD). The object dictionary consists of
a group of predefined, numbered objects with various functions.
Objects
Properties of an object:
■ An object can be viewed as a type of container. This container can simply hold data that can be read and written,
or it can possess additional functionality, so that access triggers a predefined action.
■ An object can be addressed via a unique 16-bit index.
■ An object can contain up to 256 subobjects, each addressed via an 8-bit subindex.
■ An object is defined by a number of parameters, including access rights, data type and type of data transfer.

14 Due to the differences in the physical layer and the fixed real-time behavior of POWERLINK compared to CAN bus physics, there are minor differences between CANopen and
POWERLINK.
16 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

An object can be defined with the following parameters:

Index 0x1F98

Name NMT_CycleTiming_REC

Object type RECORD

Data type NMT_CycleTiming_TYPE

Category 15.

Table 2: Parameters of an object

Subindex 0x03

Name PResMaxLatency_U32

Data type UNSIGNED32 Category CN: M, MN: -

Value range UNSIGNED32 Access const16

Default value no PDO mapping No

Table 3: Parameters of a sub-object

Structure of the object dictionary

The object dictionary is divided into multiple ranges, which are described in Tab. 4. Most important for POWERLINK
users is the object range ≥ 0x2000. POWERLINK technology integrators can define objects freely in this range.
Index Object range description
0x0000 Not used

0x0001 .. 0x001F Static data types

0x0020 .. 0x003F Complex data types

0x0040 .. 0x005F Manufacturer-specific complex data types

0x0060 .. 0x007F Device profile specific static data types

0x0080 .. 0x009F Device profile specific complex data types

0x00A0 .. 0x03FF Reserved for further use

0x0400 .. 0x041F POWERLINK-specific static data types

0x0420 .. 0x04FF POWERLINK-specific complex data types

0x0500 .. 0x0FFF Reserved for further use

0x1000 .. 0x1FFF Communication profile area

0x2000 .. 0x5FFF Manufacturer-specific profile area

0x6000 .. 0x9FFF Standardized device profile area

0xA000 .. 0xBFFF Standardized interface profile area

Table 4: "Object dictionary structure" from EPSG DS 301 V1.x.x - Communication Profile Specification

Application data that is transferred cyclically or acyclically is found the object range ≥ 0x2000.

15 MM: Mandatory; object must exist

16 const: Constant; value can only be read and does not change.
 POWERLINK – TECHNOLOGY 17

Device profiles
Object groups in the range ≥ 0x6000 have been predefined in order to provide a consistent, vendor-independent in-
terface for specific device types (see Tab. 4). A group of objects dedicated to a specific device type is called a device
profile. Some examples include rotary encoders, frequency inverters and valve controllers.
It is not absolutely necessary to use one of these standard interfaces, but for components that are not permanently
integrated in an overall system it is recommended.

Standardized device profiles are provided by the user organization CAN in Automation (CiA)17. This orga-
nization also assigns each device a CANopen Vendor IDT18.
The Ethernet POWERLINK Standardization Group (EPSG)19 can also assign a (CANopen) vendor ID. How-
ever, this is then only permitted be used for POWERLINK devices.

3.3.2 Service Data Objects (SDO)

The Service Data Object protocol can provide asynchronous access to objects. The object dictionary of a POWERLINK
node can be accessed from any other node using the SDO protocol as long as the object exists and access is permitted.
The SDO protocol operates according to the server/client principle. The client reads/writes data stored on the server.

Usage Example
Non-cyclic data access Transferring object data ≥ 0x2000 in the asynchronous phase (ASnd
frame)

POWERLINK network configuration Startup of a POWERLINK node

Diagnostics Read error counter

File transfer Firmware upgrade

Table 5: Possible uses of the SDO protocol

17 www.can-cia.org

18 he Vendor ID identifies the manufacturer of a CANopen or POWERLINK device.

19 www.ethernet-powerlink.org
18 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

The server20 contains e.g. data in object 0x1C0B/0x01. Another POWERLINK node (SDO client) can access
this data by sending an SDO read command (containing the index and subindex) to the SDO server. The
server answers with an SDO response containing the data from object 0x1C0B/0x01 (in example 0x06).

Figure 12: SDO server-client principle

SDO communication occurs in the asynchronous phase, i.e. the SDO frame is embedded in the ASnd
frame.
SDO read or write commands can also be bundled21 to send multiple read or write commands in a single
ASnd frame. This can increase the throughput of SDO communication.

3.3.3 Process data objects (PDO)

Cyclically transferred process data objects are embedded in PDO frames. A PDO frame can contain one or more objects
and their data.
Transfer PDO properties:
■ PDO frames transferred cyclically, i.e. time-based not event-based
■ PDO frames are sent as user data in the PRes or PReq frame (see "POWERLINK Payload" in Fig. 5 "POWERLINK
frame structure")
■ Producer-consumer principle:
■ Every node can produce object data for the network in the form of PDO frames.
■ One or more other nodes can read this PDO frame in the same cycle.
■ There is no direct connection between any two nodes.

20 Any particular POWERLINK node

21 Using "Read Multiple Parameter by Index" or "Write Multiple Parameter by Index"

 POWERLINK – TECHNOLOGY 19

TPDO and RPDO

A PDO frame is sent by a network station (producer). For this node, the frame is a Transmit PDO (TPDO) frame. For
the node (consumer) that receives it, this frame is a Receive PDO (RPDO) frame. In other words, any PDO frame can
be seen as either a TPDO or RPDO depending on the perspective (see Fig. 14 "TPDO mapping of an object" on page
21 and Fig. 15 "RPDO mapping of an object").

3.3.4 Mapping

How does user data get from the network (from a PDO frame) into a process variable22 and vice versa?
The answer to this question is summarized as the term "mapping". Mapping functionality is derived from POWERLINK's
CANopen application layer.

Mapping is the process of assigning cyclic network data to a specific node's objects (and their respective
data points / process variables). To exchange cyclic data between two nodes, then, mapping (as well as
source and target objects) is required independently on both the transmitter and receiver ends.
With POWERLINK, mapping occurs in mapping objects.

Mapping objects
Mapping is defined by objects contained, like all other objects, in a bus station's object dictionary. Mapping objects
can be found on both the MN and CNs.
The object dictionary has a range dedicated specifically to mapping.
Object area Object name Contents
RPDO23 0x1400 to 0x14FF RPDO communication parameter Mapping version and source ad-
dressing

0x1600 to 0x16FF RPDO mapping parameter "RPDO to object" connections

TPDO24 0x1800 to 0x18FF TPDO communication parameter Mapping version and target ad-
dressing

0x1A00 to 0x1AFF TPDO mapping parameter "Object to TPDO" connections

Table 6: Object range for PDO mapping objects.

RPDO and TPDO mapping always require a corresponding pair of communication and mapping parameters (see Tab.
6) that provides the network all the necessary information about the object data link. This pair is always identified by
the last two characters in the index number.
Example for PDO mapping object pairs (see Tab. 6):
■ RPDO: 0x1401 and 0x1601
■ TPDO: 0x1800 and 0x1A00

22 In this training module, "process variable" refers to application variables that are refreshed cyclically via POWERLINK.

23 Receive Process Data Object

24 Transmit Process Data Object

20 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Relationship between object dictionary and PDO mapping

The following image (Fig. 13) illustrates the relationship between the object dictionary, mapping objects, application
objects, process data objects (PDOs), POWERLINK frames and process images.
The mapping objects establish the correlation between the cyclic POWERLINK frame payload data (PDO frame) and
the application object or process image for each node. A concrete example of the content of a (RPDO) mapping object
and its interpretation are provided below. TPDO mapping works the same way.

PDO mapping does not create a direct data link between two nodes, but rather describes the correlation
between a data block in the PDO frame (i.e. network) and a source/target object (i.e. application data
point). This source/target object is also found in the object dictionary, just in a different range (≥ 0x2000)
than the PDO mapping objects.
Information about how PDO data is to be interpreted is not transferred along with the PDO data, but is
contained in the object dictionary.

MN CN
Application engineers Application engineers
(PLC processing) (PLC processing)

Process image Process image

(Optional) (Optional)

OD OD
Manufacturer- and Manufacturer- and
device-specific device-specific
objects objects

0x2000 0x2000
… …

PReq
TPDO RPDO
OD PRes MN OD
Mapping objects Mapping objects
... Copy Copy ...
0x14xx according to POWERLINK according to 0x14xx
0x16xx mapping mapping 0x16xx
0x18xx 0x18xx
0x1Axx RPDO PRes TPDO 0x1Axx
… …

Figure 13: The content of mapping objects defines correlation between network and application

Steps to determine the mapping

The mapping structure for one transfer direction is determined by the following steps:
 POWERLINK – TECHNOLOGY 21

1) Which nodes are communicating with each other?

=> TPDO, RPDO communication parameters of respective node
(Tab. 6 "Object range for PDO mapping objects." on page 19)
2) Which data should be transferred?
=> Source/target objects in node's respective manufacturer-/device-specific object range (Tab. 4 ""Object dictio-
nary structure" from EPSG DS 301 V1.x.x - Communication Profile Specification" on page 16)
3) Where is the object data in the PDO?
■ Node sending data:
=> TPDO mapping parameters (Tab. 6), source objects sent together in TPDO
■ Node receiving data:
=> RPDO mapping parameters (Tab. 6), RPDO data copied to target object

These steps represent a pattern that can be applied to the examples below with the same numbering.

3.3.5 Mapping example: Rotary encoder sends position data to the MN

1. Transfer direction: Rotary encoder (CN1) to network - Definition of TPDO mapping

1) Which nodes are communicating with each other?

=> Encoder (CN1) sends data to the MN. A CN can only ever
CN send cyclic PDOs to all POWERLINK stations.
Application 1 Application
Software VarPosition [U16] 2) Which data should be transferred?
=> Position data of the encoder, i.e. data of the source ob-
ject 0x2001/0x43 with the size 16 bits (0x10).
OD
Manufacturer- and 3) Node sending data (encoder): Where should the source ob-
device-specific 0x2001/0x43 [U16] ject data be located in the TPDO?
Object Objects
=> Right at the beginning of the TPDO with offset 0
Dictionary
Copy
OD according to
Mapping 0x1800/0x01
mapping
Objects 0x1A00/0x01

Payload [U16]

Network
header TPDO Frame CRC
PRes

Figure 14: TPDO mapping of an object

Questions 1 through 3 determine the content of the mapping objects for the rotary encoder. According to the POW-
ERLINK specification for CNs, the answer to Question 1 is always a target node ID of 0x00. The content of object
0x1A00/0x01 is composed of the answers to Questions 2 and 3. Objects 0x1800 and 0x1A00 form a pair that represents
a unique description of the TPDO - see Tab. 6 "Object range for PDO mapping objects." on page 19.
Result of question Object Contents Meaning
no.
1 0x1800/0x01 0x00 Target NodeID: Always 0 for
CN

2 and 3 0x1A00/0x01 0x0010000000432001 See following breakdown

Table 7: Mapping objects of rotary encoder (CN1)

0x0010 0x0000 0x00 0x43 0x2001

Length in bits Offset in bits Reserved Subindex source ob- Index source object
ject

Table 8: Breakdown of object 0x1A00/0x01

22 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

2. Transfer direction: Network to MN – Defining the RPDO mapping

Based on the PDO mapping of the transmitted frame defined above, we can now define the RPDO mapping on the
receiving end.
1) Which nodes are communicating with each other?
=> Encoder (CN1) sends data to the MN. In other words, the
MN MN receives a PDO from the network with a source node ID
Application PLC application
Software VarPosition [U16] of 1.
2) Which data should be transferred?
=> Position data of the encoder; stored on the MN in target
OD
Manufacturer- and object 0x3200/0x05 with a size of 16 bits.
device-specific 0x3200/0x05 [U16]
3) Node receiving data (MN): Where in the RPDO is the target
Object Objects
object data located?
Dictionary
=> Position data of the encoder is located in the PDO at the
OD Copy
Mapping 0x1400/0x01 according to position defined by the previous TPDO mapping, i.e. right at
Objects 0x1.600/0x01 mapping the beginning of the RPDO with offset 0.

Payload [U16]

Network
header RPDO Frame CRC
PRes

Figure 15: RPDO mapping of an object

The answers to questions 1 through 3 determine the content of the mapping objects for the MN. A 0x16xx subobject
in the MN's object dictionary defines how it is to interpret the RPDO data - see Tab. 6 "Object range for PDO mapping
objects." on page 19. The content of this subobject depends on the structure of the RPDO.
The CN that the MN receives data from is determined by the associated object (object pair) of 0x1600: PDO commu-
nication parameter 0x1400. The content of object 0x1400/0x01 reflects NodeID 1 of the rotary encoder since this is
the node from which the PDO is received.
Result of question Object Contents Meaning
no.
1 0x1400/0x01 0x01 Source NodeID: 1 (encoder)

2 and 3 0x1600/0x01 0x0010000000053200 See following breakdown

Table 9: Mapping objects of rotary encoder (CN1)

0x0010 0x0000 0x00 0x05 0x3200

Length in bits Offset in bits Reserved Subindex target ob- Index target object
ject

Table 10: Breakdown of object 0x1600/0x01

 POWERLINK – TECHNOLOGY 23

3.3.6 Mapping example: Rotary encoder sending position data directly to an X20
bus controller
1. Transfer direction: Rotary encoder (CN1) to network - Definition of TPDO mapping

1) Which nodes are communicating with each other?

CN
=> Encoder (CN1) also sends data to the X20 bus controller
Application 1 Application
Software VarPosition [U16] VarSpeed [U32]
(CN2). A CN can only ever send cyclic PDOs to all POWERLINK
stations.
OD 2) Which data should be transferred?
Manufacturer- and
device-specific 0x2001/0x43 [U16] 0x2002/0x01 [U32] => Velocity data of the encoder, i.e. data of the source ob-
Object Objects
ject 0x2002/0x01 with the size 32 bits (0x20).
Dictionary
OD Copy 3) Node sending data (encoder): Where should the source ob-
Mapping 0x1800/0x01 according to Copy
Objects 0x1A00/0x01 mapping according to ject data be located in the TPDO?
0x1A00/0x02 mapping
=> Following the position data, i.e. offset is 16 bits.
Payload [U16] Payload [U32]

Network
header TPDO Frame CRC
PRes

Figure 16: TPDO mapping of two objects

Questions 1 through 3 determine the content of the mapping objects for the rotary encoder.
Result of question Object Contents Meaning
no.
1 0x1800/0x01 0x00 Target NodeID: Always 0 for
CN

- 0x1A00/0x01 0x0010000000432001 See breakdown above (Tab. 8)

2 and 3 0x1A00/0x02 0x0020001000012002 See following breakdown

Table 11: Mapping objects of rotary encoder (CN1)

0x0020 0x0010 0x00 0x01 0x2002

Length in bits Offset in bits Reserved Subindex source ob- Index source object
ject

Table 12: Breakdown of object 0x1A00/0x02

2. Transfer direction: Network to X20 bus controller - Definition of RPDO mapping

Based on the PDO mapping of the transmitted frame defined above, we can now define the RPDO mapping on the
receiving end.
24 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

1) Which nodes are communicating with each other?

=> Encoder (CN1) sends data to the X20 bus controller
CN (CN2). In other words, CN2 receives a PDO from the network
Application 2 PLC application
Software VarSpeed [U32] with a source node ID of 1.
2) Which data should be transferred?
=> Velocity data of the encoder; stored on CN2 in target ob-
OD
Manufacturer- and ject 0x2000/0x01 with a size of 32 bits (0x20).
device-specific 0x2000/0x01 [U32]
3) Receive node (X20 bus controller): Where is the data of the
Object Objects
target objects located in the RPDO?
Dictionary
=> Velocity data of the encoder is located in the PDO at the
OD Copy
Mapping 0x1401/0x01 according to position defined by the previous TPDO mapping, i.e. in the
Objects 0x1601/0x01 mapping RPDO after the position data with a 16-bit offset (0x10).

Payload [U16] Payload [U32]

Network
header RPDO Frame CRC
PRes

Figure 17: RPDO mapping of one object with cross-traffic

The answers to questions 1 through 3 determine the content of the mapping objects for the X20 bus controller (CN2).
A 0x16xx subobject in the X20 bus controller's object dictionary defines how it is to interpret the RPDO data - see Tab.
6 "Object range for PDO mapping objects." on page 19. The content of this subobject depends on the structure
of the RPDO.
The CN that the MN receives data from is determined by the associated object (object pair) of 0x1601: PDO communi-
cation parameter 0x1401. The content of object 0x1401/0x01 reflects NodeID 1 of the rotary encoder since this is the
node from which the PDO is received.
Result of question Object Contents Meaning
no.
1 0x1401/0x01 0x01 Source NodeID: 1 (encoder)

2 and 3 0x1601/0x01 0x0020001000012000 See following breakdown

Table 13: Mapping objects of rotary encoder (CN1)

0x0020 0x0010 0x00 0x01 0x2000

Length in bits Offset in bits Reserved Subindex target ob- Index target object
ject

Table 14: Breakdown of object 0x1601/0x01

Exercise: Mapping questions

1) Which range of the object dictionary is used for mapping?

2) How many objects are required to describe a mapping?

3) How many object dictionaries are required for a cyclic data link with two nodes?

4) How can a mapping describe an acyclic data link?

5) What link is defined by the mapping?

6) How can one determine the direction of a transmission (transmitting or receiving)?

7) What could be the reason that object 0x1401 is used in the second mapping example rather than object 0x1400?

8) Object 0x1800/0xF0 contains the following value: 0x0030002000FF4000. What does this value mean?
 POWERLINK – TECHNOLOGY 25

Additional information about the mapping can be found in the EPSG DS 301 V1.x.x - Communication Pro-
file Specification / Search for objects from Tab. 6 "Object range for PDO mapping objects." on page 19

3.4 Network management (NMT)

The term "network management" (NMT) refers to all mechanisms used to query and modify the operating modes of
POWERLINK nodes. It also includes the identification of new nodes.

3.4.1 MN and CN state machines

The MN is responsible for network management (NMT). The MN controls the states of the CNs by querying status
information and issuing NMT commands. Each node follows one of the nodes described in the "EPSG DS 301 V1.x.x -
Communication Profile Specification25 (see Fig. 18 "MN state machine" and CN state machine), which describes possi-
ble state transitions. For each node there is a defined startup sequence that ends in the OPERATIONAL state.

The OPERATIONAL state indicates successful configuration and integration in the POWERLINK network.
Only when this state has been achieved does the node begin exchanging valid cyclic data.

25 " www.ethernet-powerlink.org
26 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

NMT_GS_
INITIALIZATION

NMT_MS (NMT_MT1)
NMT MN state machine auto
[NodeID ==C_ADR_MN_DEF_NODE _ID]

NMT_MS_ (NMT_MT7) NMT_MS_

NOT_ACTIVE Timeout( SoC, SoA) BASIC_Ethernet
[NMT_StartUp_U32.bit 13 == 1]

NMT_MS_EPL_MODE
(NMT_MT2)
Timeout (SoC, SoA)
[NMT_StartUp_U32.bit 13 == 0]

NMT_MS_ (NMT_MT6)
PRE_OPERATIONAL_1 Error condition

(NMT_MT3)
auto

NMT_MS_
PRE_OPERATIONAL_2

(NMT_MT4)
auto

NMT_MS_
READY_TO_OPERATE

( NMT_MT5)
auto

NMT_MS_
OPERATIONAL

Type of communication
POWERLINK POWERLINK cycle
No communication Legacy Ethernet
Listen only Reduced cycle (SoC PReq PRes
(IP and others)
(SoA ASnd) SoA Asnd)

Figure 18: MN state machine

 POWERLINK – TECHNOLOGY 27

NMT_GS_
INITIALIZATION

NMT_CS
NMT CN state machine
(NMT_CT1)
auto,
[NodeID !=
(NMT_CT3)
C_ADR_MN_DEF _NODE _ID ]
Timeout
(SoA,SoC ,PReq,PRes )

NMT_CS_
BASIC _Ethernet NMT_CS_
NOT_ACTIVE

(NMT_CT12)
SoA ,SoC,PReq,PRes
(NMT_CT2)
SoA ,SoC

NMT_CS_EPL_MODE

NMT_CS_ (NMT_CT11)
PRE_OPERATIONAL _1 Error condition

(NMT_CT4)
(NMT_CT5)
SoC
NMT
EnableReadyToOperate

NMT_CS_
PRE_OPERATIONAL _2

(NMT_CT10)
NMT
(NMT_CT6)
auto EnterPreoperational 2

NMT_CS _
READY_TO_OPERATE

(NMT_CT9)
( NMT_CT7)
NMT
NMT
EnterPreoperational 2
StartNode

NMT_CS _
OPERATIONAL

(NMT_CT8)
NMT_CS _
NMT
STOPPED
StopNode

Type of communication

POWERLINK POWERLINK cycle

Legacy Ethernet
No communication Listen only Reduced cycle (SoC PReq PRes SoA
(IP and others )
(SoA Asnd) Asnd)

Figure 19: CN state machine

3.4.2 Network startup

How startup of a POWERLINK network can be carried out step-by-step is shown as an example in Fig. 20 "Startup
procedure for a CN"26. This illustrates the state transitions of a CN (see Fig. 19 "CN state machine") as triggered by:
■ A received POWERLINK frame (SoA, SoC), i.e. status change to:
- NMT_CS_PREOPERATIONAL_1
- NMT_CS_PREOPERATIONAL_2
■ An NMT command directly addressed to the CN (contained in an ASnd frame)
- NMTEnableReadyToOperate: Change to NMT_CS_READY_TO_OPERATE
- NMTStartNode: Change to NMT_CS_OPERATIONAL

26 from EPSG DS 301 V1.x.x - Communication Profile Specification

28 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

MN CN

NMT_GS_POWERED

NMT_GS_INITIALIZING

NMT_GS_RESET_APPLICATION
NMT_GS_RESET_CONFIGURATION

NMT_MS_NOT_ACTIVE NMT_CS_NOT _ACTIVE

NMT_MS_PRE_OPERATIONAL _1 NMT_CS_BASIC_Ethernet _MODE

Start reduced cycle SoA

NMT_CS_PRE_OPERATIONAL _1

SoA (R. ServiceID : IdentRequest ) SoA BOOT_STEP1

ASnd ASnd (ServiceID : IdentResponse )

SDO communication [Client] *

*
SDO communication [Server]
SoA (R. ServiceID : IdentRequest ) SoA

ASnd ASnd (ServiceID : IdentResponse )

BOOT_STEP1 E_OK

NMT_MS_PRE_OPERATIONAL _2

Start isochronous cycle SoC

NMT_CS_PRE_OPERATIONAL _2

Command : NMTEnableReadyToOperate ASnd BOOT_STEP2

NMT_CS_READY _TO_OPERATE

SoA (R. ServiceID : StatusRequest ) SoA

ASnd ASnd (ServiceID : StatusResponse )

BOOT_STEP2 E_OK

NMT_MS_READY_TO_OPERATE

Send PReq PReq CHECK_COMMUNICATION

PRes
Send PRes

CHECK_COMMUNICATION E _OK

NMT_MS_OPERATIONAL

Command : NMTStartNode ASnd START

NMT_CS_OPERATIONAL

SoA (R. ServiceID : StatusRequest ) SoA

ASnd ASnd (ServiceID : StatusResponse )

START _CN E_OK

Operational

* Operation is done several times

Figure 20: Startup procedure for a CN
 POWERLINK – DIAGNOSTICS 29

4 POWERLINK – Diagnostics
4.1 Diagnostics tools

Often it is no simple task to figure out why a machine won't start – or even worse, why it stopped working in the middle
of a process. When it comes to pinpointing the source of a malfunction as quickly as possible, there's no substitute for
powerful diagnostics tools and the knowledge of how to use them. These tools are presented in the following sections.

4.1.1 Automation Studio

Automation Studio is the first place to go to identify malfunctions in a POWERLINK network. The status data points
and the logger provide a quick initial diagnosis.
ModuleOK
The I/O mapping contains a data point called "ModuleOK"27. For POWERLINK nodes (e.g. X20 bus controller) it is set
to TRUE, as long as:
■ The CN is in the state OPERATIONAL
■ The RD flag in the CN's PRes frame has the value "1".
This means that if the frame contains PDO data, then it is valid.

Figure 21: ModuleOK flag in I/O mapping

Module supervised
By default, the MN monitors the Automation Studio hardware configuration for any nodes added to the POWERLINK
network. This can be configured on the CN28 using the "Module Supervised" option. When "Module Supervised = On",
the PLC will enter service mode if the respective CN fails.

Figure 22: CN's "Module Supervised" parameter in the Configuration View

Logger
Any network errors detected by the MN, such as problems with SDO communication, are also entered in the logger.

Figure 23: Logger reports a failed SDO write access to an object of CN1 (IF3.ST1).

27 If "ModuleOK" is set to FALSE, then the node is either not in the OPERATIONAL state or its data is invalid – it does not necessarily mean that there is anything wrong with the
module itself.

28 The same setting is available for individual X20 modules and has the same effect.
30 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Activating the fieldbus module view in the logger

■ POWERLINK-relevant errors are also displayed in the logger's fieldbus module view.
This view must be activated separately.

Figure 24: Logger module window

To look up the meaning of the error codes, you can either select the error in the logger and press F1, or navigate to
the corresponding section of Automation Help:

Communication / POWERLINK / Diagnostics / Error numbers

Diagnostic data points

In addition to "ModuleOK" there are other diagnostic data points that can be helpful in locating errors.

The image on the left shows the diagnostic data

points available for an X20BC0083 bus controller:
■ For example, if the POWERLINK connection on
the bus controller is lost (and reestablished),
the "EthPhyXLinkLoss" counter is incremented.
■ A short description of the data points can be
found in Automation Studio in the correspond-
ing column of the I/O mapping window .
■ The diagnostic data points that monitor an
Ethernet interface (hence the "Eth" at the be-
ginning of their name) are read asynchronously
via the SDO protocol.

Figure 25: Diagnostic data points for the X20BC0083 in the I/O mapping window
(monitor mode)

The Ethernet connection and corresponding cabling of POWERLINK stations can be monitored using the
predefined diagnostics data points for B&R devices – either from the MN or from Automation Studio.
This speeds up the process of pinpointing the source of errors at a hardware level.

Communication / POWERLINK / Diagnostics / Diagnostic data points

SDO library - AsEPL

Automation Studio provides a library that can be used, for example, to access a CN's objects using the SDO protocol.
This feature can also be used for diagnostic purposes by reading the contents of an object to check a device's status
or an error code. In order to do so, you need to know the index and subindex of the object.

Programming / Libraries / Communication / AsEPL

 POWERLINK – DIAGNOSTICS 31

Relevant diagnostics objects are:

■ 0x1003: "ERR_History_ADOM"29:
■ 0x1C00 to 0x1C17: "DLL error handling objects"

Further information
For more information about the features described here, as well as additional diagnostic options, consult Automation
Help:

Diagnostics and service / I/O and network diagnostics / POWERLINK

Communication / POWERLINK / Diagnostics

The following training modules provide helpful information about diagnostics in general (independently of POWER-
LINK):
■ TM920 – Diagnostics and Service for End Users (independently of Automation Studio)
■ TM223 - Automation Studio diagnostics

Exercise: Create an Automation Studio project based on the hardware in the ETAL950 cube.
In the following exercise, a project will be created that will later be used to illustrate various concepts.

1) Create an Automation Studio project and name it "POWERLINK_ETA" (to be used later) with the following hard-
ware configuration:

Figure 26: Overview of hardware configuration for the "POWERLINK_ETA" project

■ X20 CPU (PLC)

■ X20BC0083 with NodeID 1 + X2X module (power supply, analog input)
■ X20BC0083 with Node ID 2 + X2X module (power supply, digital input, analog output)
■ "Module supervised = off" for all modules

2) Create a program named "CopyAItoAO" as shown in Tab. 18 "ANSI C program "CopyAItoAO"" on page 59 with
the corresponding variable declaration (local, same data type as individual analog channel, i.e. "INT").

3) Assign the "CopyAItoAO" program to the task class "Cyclic #1" .

29 Not all nodes support all of the objects (i.e. optional objects) specified in "EPSG DS 301 V1.x.x - Communication Profile Specification".
32 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

4) Map the variables of the "CopyAItoAO" program to the AI and AO modules as shown in Fig. 48 "CN to CN connec-
tion via the MN" on page 59.

5) Set up a direct CN-to-CN connection of CN1 data point "AnalogInput01" and CN2 data point "AnalogOutput01",
as shown in Fig. 49 "Direct CN to CN connection" on page 60. This connection is referred to as cross-traffic or
"crosslink" and doesn't require any user variables or programs on the PLC.
See references to Automation Help provided in section 5.4.1.3 "Direct cross-traffic" on page 58.

The ETA cube's control panel behaves as described in Tab. 19 "Description of POWERLINK ETA test system
functions" on page 61. Turning a potentiometer (P1, P2) changes the value of the 7-segment display
from 0 to 9.9. P1 modifies the left display, and P2 the right display.

4.1.2 System Diagnostics Manager

Most of the diagnostics options pre-

sented in section 4.1.1 "Automation
Studio" are also available in System Di-
agnostics Manager.
This allows the majority of diagnostic
tasks to be performed independently
of Automation Studio.

Figure 27: System Diagnostics Manager view in browser window

Diagnostics and service / Diagnostics tools / System Diagnostics Manager

Exercise: Read objects using the SDO protocol.

Use the AsEPL library to perform asynchronous read access to an object on the CN. Control this access in Automation
Studio using the Watch window. Read access to the network occurs as shown in Fig. 12 "SDO server-client principle"
on page 18. In this case, the bus controller (CN) is the SDO server.

1) Open the Automation Studio project "POWERLINK_ETA".

2) Add a sample implementation of the AsEPL library.

 POWERLINK – DIAGNOSTICS 33

Figure 28: Add an AsEPL sample implementation in the Logical View.

3) Modify the eplCyclic.st file for the sample implementation. In the step EPL_READ_OE, change the following lines
(changes in red):

EPL_READ_OE: (*state to read an object entry*)

epl.fub.EplSDORead_0.enable := 1;
(*enable the FUB*)
epl.fub.EplSDORead_0.pDevice := ADR('IF3');
(*Device Name of Powerlink interface*)
epl.fub.EplSDORead_0.node := 2;
(*Node Number of the Station to be read*)
epl.fub.EplSDORead_0.index := 16#1C0B;
(*Read Index 16#1C0B : CNLoss SoC*)
epl.fub.EplSDORead_0.subindex := 1;
(*Subindex 1 is the CumulativeCnt_U32*)
epl.fub.EplSDORead_0.pData := ADR(epl.para.readData);
(*Pointer to the variable for the read Data*)
epl.fub.EplSDORead_0.datalen := SIZEOF(epl.para.readData);
(*Length of the Data to be read*)
4) Transfer project to the PLC30

30 Program Compact Flash card if necessary

34 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

5) Open the Watch window for the object "eplCyclic.st".

6) Add the variable "epl".

7) Set "epl.cmd.readThresholdCnt" to "TRUE".

8) The value read from bus controller object 0x1C0B/0x0131, which counts the number of lost SoC frames, appears
in "epl.para.readData".

9) Cause a brief interruption of POWERLINK communication on Node 232 (i.e. disconnect/reconnect Ethernet cable).

10) Read object 0x1C0B/0x01 on Node 2 again.

11) Open System Diagnostics Manager and check the status variables33 of the X20BC0083 module with NodeID 2:
One of the error counters (EthPhyXLinkLoss) should match the exact number of interruptions.
Nevertheless, this counter is only conditionally comparable with the 0x1C0B/0x01 object, which doesn't count
the number of connection interruptions, but the number of failed SoC frames. Both counters are incremented in
the event of interruption, however.

The CN has an internal time monitor of received SoC frames. When a frame is lost, the value of the
0x1C0B/0x01 object is incremented. The MN can read this value using the SDO protocol (3.3.2 "Service
Data Objects (SDO)" on page 17).
Reading error counter objects like this (see "SDO library - AsEPL" on page 30) can also help identify
sporadic interruptions of the Ethernet connection and localize them using the node number.
B&R devices offer additional error counters34, that are very helpful for diagnosing physical network con-
nections (Ethernet cable and connections). These counters can be read automatically via SDO and viewed
in Automation Studio or System Diagnostics Manager.

4.1.4 Overview of external diagnostic tools

Additional diagnostics options are available in the form of external diagnostic tools. Since POWERLINK is based on
standard Ethernet, all Ethernet-based network diagnostics tools can be used. For diagnosis of difficult network prob-
lems, the following solutions have proven effective:
■ WireShark (free - GPL35): http://www.wireshark.org/
■ B&R Analyzer for Wireshark TM: At www.br-automation.com search for "Wireshark

31 see "EPSG DS 301 V1.x.x - Communication Profile Specification": Object "DLL_CNLossSoC_REC"

32 Make sure that the node and its I/O modules are not supervised in Automation Studio (Module supervised = off).

33 see Fig. 25 "Diagnostic data points for the X20BC0083 in the I/O mapping window (monitor mode)" on page 30

34 Not all B&R devices provide these counters (see Automation Help under "Diagnostic data points").

35 GPL: General Public License; changes to the software must also be released under the GPL.
 POWERLINK – DIAGNOSTICS 35

These programs allow you to log and analyze network traffic using a PC and a standard network card. Wireshark comes
with POWERLINK support. For additional analysis options, you can download a plug-in from the B&R website. Simply
search the B&R website for "Wireshark" and download the "B&R Analyzer for Wireshark TM" plug-in. This plug-in in-
cludes, among other things, logical analysis of network traffic with advanced statistics, as well as the possibility to
connect to hardware for accurate timestamps (X20ET8819).
In this manual, only the use of Wireshark will be discussed in detail.

4.1.5 Wireshark

This section provides an introduction to working with the "Wireshark" tool for capturing and analyzing network com-
munication.
Requirements for network recordings

It is a good idea to install a second Ethernet card in the PC that will be performing the capture, since the
primary one is generally needed for other purposes, such as connecting to the PLC.

To record data traffic, an Ethernet port on the PC performing the capture is connected directly to any station in the
POWERLINK network. Ideally this is done using the same type of cable36 that is used for the POWERLINK network itself.
The location on the network that the packets are captured has only a minor effect on the timing between the frames
(hub and tolerance times, etc.) and no effect on their order.

Figure 29: POWERLINK network analysis using a PC

All protocols (TCP/IP, Windows networks, etc.) should be deactivated in the control panel on the PC per-
forming the capture to prevent unwanted packets from disturbing the POWERLINK network.
Exception: The Npcap Packet driver (NPCAP) must remain enabled; otherwise, the interface will not ap-
pear in Wireshark.

Most network cards have one or more LEDs to indicate the connection status and network activity. As soon as the
connection to an active POWERLINK network is established, these LEDs should light up accordingly.

If the PC fails to connect to the POWERLINK network, check the advanced settings of the PC's network
card. A hub can also be connected between the PC and the network, since the physical connection is
already properly preconfigured and does not require any settings to be changed.
Recommended network card settings:
■ Half-duplex
■ 100 Mbit/s transfer rate
■ Auto crossover (optional)

36 Cable recommendations can be found in section 2.2 "POWERLINK" on page 6.

36 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Wireshark installation
The download link to the latest version and information about the Wireshark installation can be found at:
• http://www.wireshark.org/

Starting a POWERLINK network recording

Once the Wireshark window has opened there are a number of ways to begin a capture. One of them is about:
1) Open "Interface list"
2) Select the network interface with a POWERLINK connection
3) Press "Start" button

Figure 30: Starting a network capture with Wireshark

This opens the "Capture Interfaces" window, showing you the activity on the POWERLINK network. If the packets are
sorted correctly ("Time" column in ascending order) then the recurring POWERLINK cycle can be observed.

Figure 31: POWERLINK cycle in Wireshark

The precision of the timestamp is determined by the interaction between the operating system and net-
work card on the PC performing the capture. The level of precision is generally enough to show the frames
in the correct order, but not enough to represent the exact time that they were received.
To get a reliable timestamp in high resolution you need to use special network analysis equipment. These
tools are introduced in section 4.1.6 "Hardware-supported network analysis".

Filter application
Filters can be used to gain a more focused view of network activity, showing only specific frames or connections. A
distinction is made between two types:
 POWERLINK – DIAGNOSTICS 37

■ Display filter: Only affects which frames of a currently running or saved capture are displayed.

Figure 32: Using a display filter in Wireshark

As shown above, the display filter can be entered in the field at the top with the aid of syntax. It is possible to use
logical operations (similar to C syntax). One the filter is applied, only packets that meet the requirements expressed
in the filter will be displayed. In the example above, these include all packets that are either sent from the node with
POWERLINK NodeID "1" (epl.src == 1) or sent to the node with NodeID "1" (epl.dest == 1).

■ Capture filter: Only packets that meet the criteria of the filter are captured. This can considerably reduce the size
of a saved capture. The capture filter uses a different syntax and should only be used by experts.

Exercise: Perform a network capture with Wireshark.

In this exercise, we will be performing a network capture during startup of a CN.

1) Open the Automation Studio project "POWERLINK_ETA"37 and transfer it to the PLC 38.

2) Connect the PC network card to any hub interface on the active POWERLINK network. The X20 bus controller also
has an internal hub (for two interfaces).

3) If you have problems connecting, consult section "Requirements for network recordings" on page 35 .

4) Verify the connection (LEDs on PC network card showing activity).

5) Switch off the CNs or the power supply to the POWERLINK nodes.

6) Starts the Wireshark capture – see "Starting a POWERLINK network recording" on page 36

7) Switching on the POWERLINK nodes

8) Stop capture after approx. 1 minute.

9) Save the capture in *.pcap format (e.g. under POWERLINK_bootup.pcap).

If the PC's network card is configured correctly, then Wireshark can be used to capture all POWERLINK
network traffic to be analyzed in more detail.

Exercise: Add a column to view cyclic payload data.

In this exercise, we will be adding another column to let us view cyclic POWERLINK payload data.

37 Created in "Exercise: Create an Automation Studio project based on the hardware in the ETAL950 cube." on page 31

38 Program CompactFlash card if necessary

38 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

1) Apply the display filter shown in the image to the capture performed in the previous exercise. The filter ensures
the validity of the PDO data.

2) Select a PRes frame

3) In the detailed view, right-click on the "Data" field below and select "Apply as column".

4) The new column should appear in the network capture list above.

5) Rename the new column to "PRes payload" by right-clicking on the column and selecting "Edit Column Details".

6) The column can only display values if the current capture does not contain a (dynamic) SDO mapping configu-
ration. To do this, restart the recording as soon as the controlled node is in the OPERATIONAL state in order to
then observe the data during operation - i.e. no stored recordings as previously.

7) Manipulate a DI from a DI module with an attached switch and observe the manipulation in Automation Studio
(or SDM).

8) The same modification must also be visible in the CN2's "PRes payload" in Wireshark. Follow the continuous trace
in Wireshark by activating the automatic scroll function with "Go / Auto Scroll in Live Capture" or by clicking on
the button shown below. Section "Filter application" on page 36 may be helpful here. Recommended filter:
"epl.src == 2 && epl.pres"

POWERLINK transmits payload data in Little Endian format. Keep this in mind when interpreting SDO
and PDO payload data in the Wireshark capture.
 POWERLINK – DIAGNOSTICS 39

From the detailed view of a received frame in Wireshark, it is easy to create new columns to view addi-
tional information.

4.1.6 Hardware-supported network analysis

Limitations of network capture using standard PC
Network captures obtained using a standard PC may be insufficient to effectively diagnose errors for the following
reasons:
■ The timestamps of captured frames are not precise enough.
■ The event that caused the error occurred outside of the capture time frame.
■ The capture file is too large to work with.
■ The error occurs sporadically

Network capture using B&R X20ET8819

The X20ET8819 network analysis tool overcomes these problems and provides a network capture with highly precise
and reliable timestamps. Digital inputs and filters can also be used to trigger a capture – this is especially helpful for
sporadically occurring errors. This tool can be downloaded from the B&R website along with its features and docu-
mentation.
■ www.br-automation.com /
Products / Control systems / X20 system / X20 hub systems / X20ET8819

Figure 34: Network analysis

Figure 33: Hardware-supported network recording tool X20ET8819

Network capture using netAnalyzer from Hilscher

A comparable tool called netAnalyzer is available from Hilscher:
■ http://www.hilscher.com

These tools are introduced here to round off the topic of hardware-based network analysis. More detailed
documentation and additional software requirements must be obtained from the respective manufac-
turer.

4.2 Startup behavior of a POWERLINK node

POWERLINK nodes follow a fixed scheme during startup, which - as described in 3.4 "Network management (NMT)"
on page 25 - is defined by their state machines. The green LED on a POWERLINK node indicates its current status. The
NodeID, or node number, gives the POWERLINK node a unique address. A misconfiguration has an effect right away.
For this reason, the NodeID is also a startup issue and is included in this section.
40 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

4.2.1 LED status indicators

Image LED Color Description

S/E1) Green/ For a description, see the data sheet on the B&R website:
Red
www.br-automation.com / Products / Control systems / X20 sys-
L/A IFx Green
tem / Bus controllers / X20BC0083 / Downloads / Documentation /
X20BC0083 data sheet

Table 15: X20BC0083 - Status LEDs

1) LED "Status/Error" is a green/red dual LED.

Every POWERLINK device uses the same LED status/error indicators and corresponding blink codes. A general descrip-
tion of the LEDs and their behavior is provided in EPSG DS 301 V1.x.x - Communication Profile Specification.

Exercise: Observe the CN's LED status indicators.

Use the LEDs to check the network status of a CN.

1) Open the data sheet for the X20BC0083 module (Tab. 15 "X20BC0083 - Status LEDs" on page 40)

2) Switch off the MN (PLC) and CN1.

3) For the following steps:

■ Observe the "S/E" and "L/A" LEDs on CN1.

■ Make a note of the state changes that occur based on the blink codes.

4) Switch on the MN and CN (at the same time).

5) Turn the CN off and then on again.

6) Disconnect the MN from the network and then reconnect it.

7) Switch off the MN or disconnect the CN from the network (same effect).
 POWERLINK – DIAGNOSTICS 41

Observations for each step in the exercise:

3) With the CN is switched on, it enters BASIC_ETHERNET mode, because no POWERLINK communica-
tion is detected on the network. After that, status changes to PRE_OPERATIONAL_1 (very short sin-
gle flash), PRE_OPERATIONAL_2 (double flash) and finally OPERATIONAL39 (On) can be seen.
4) Similar behavior, except that the CN seems40 to switch straight to PRE_OPERATIONAL_2.
5) The red blinking LED indicates a faulty network connection. "L/A" indicates that there is a link, but
no activity.
6) "L/A" LED now shows neither link nor activity.

All in all, just a quick glance at the LED status indicators is needed to recognize which status the node is in.
These 3 cases should be distinguishable:
• Light on = OPERATIONAL (ready for use)
• Light blinking = PRE_OPERATIONAL_1, PRE_OPERATIONAL_2 or READY_TO_OPERATE (node is being
configured)
• Flickering = BASIC_ETHERNET (no connection to MN)

If the LED status indicators continue to blink, resort to more in-depth diagnostics (e.g. with Wireshark).

Exercise: Observe the MN's LED status indicators.

Use the LEDs to check the network status of an MN.

1) Open the X20CP1586 data sheet (go to www.br-automation.com and search for "X20CP1586")
Note: The blink codes behave largely the same as the ones for the X20BC0083.

2) Switch off the PLC and disconnect from the POWERLINK network.

3) For the following steps:

■ Observe the "S/E" LED on the MN (PLC).

■ Make a note of the state changes that occur based on the blink codes.

4) Switch on MN again.

5) Compare your observations with the CN's "S/E" behavior in the previous exercise.

The "S/E" LED on the MN (PLC) behaves the same as the one on the CN. However, the MN starts up
independently, even without a network connection.

Different blink speeds indicate different POWERLINK states. With the exception of the BASIC_ETHERNET
state (rapid flickering), the following rule of thumb applies: The faster the blinking frequency, the higher
the state (LED on = OPERATIONAL). The state of the CN will always trail behind the state of the MN. In
other words, the CN can never be in a higher state than the MN.

Exercise: Interpreting a network recording

Examine the startup sequence of a POWERLINK CN using a network capture.

1) Open the *.pcap file from "Exercise: Perform a network capture with Wireshark." on page 37

2) Apply and combine (logical operations using ==, !=, &&, ||) various display filters.

39 The READY_TO_OPERATE status is too short to be observed

40 PRE_OPERATIONAL_1 is too short to be observed

42 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Meaning/Application Filters

• Startup of all nodes epl.asnd.svid

• NMT commands and SDO traffic

Start up a CN with the node number <CN NodeID> (epl.src == <CN NodeID> || epl.dest == <CN
NodeID>) && (epl.asnd.svid || epl.pres.stat
>= 0x6d)

SoA frame contains IdentRequest epl.soa.svid == 1

SoA frame contains StatusRequest epl.soa.svid == 2

ASnd frame contains IdentResponse epl.asnd.svid == 1

ASnd frame contains StatusResponse epl.asnd.svid == 2

ASnd frame contains StatusResponse with PRE_OPERATIONAL_1 epl.asnd.sres.stat == 0x1d

ASnd frame contains SDO epl.asnd.svid == 5

PRes with status OPERATIONAL epl.pres.stat == 0xfd

Table 16: Wireshark display filters for CN startup

3) Use the search function (keyboard shortcut "CTRL+F") to find the recording of any character string in the high-
lighted text (e.g. an NMTReset command).

4) Find another filter for the NMT command "NMTEnableReadyToOperate" in the POWERLINK specification.
The specification can be downloaded from the B&R website under:
www.ethernet-powerlink.org / Downloads / Technical Documents / EPSG DS 301 V1.x.x - Communication Profile
Specification

5) Compare the startup sequence to Fig. 20 "Startup procedure for a CN" on page 28. Use a suitable filter from Tab.
16 to do this.

The startup sequence of a POWERLINK CN can be observed using Wireshark. If you are familiar with the
steps of the startup sequence, this method is ideal for troubleshooting a node that fails to reach the
OPERATIONAL state.
 POWERLINK – DIAGNOSTICS 43

4.2.4 POWERLINK node number

Every POWERLINK node has a node number, or NodeID, that uniquely identifies it in the POWERLINK segment. For easy
maintenance41 this number is generally set using two hexadecimal rotary switches.

The node number is applied when the station is switched on. Any modifications made after startup will
be applied after the next startup procedure. The position of the node number switches must match the
configuration of the respective POWERLINK node in Automation Studio.

Communication / POWERLINK / AR configuration / Insert module

Communication / POWERLINK / Addressing

POWERLINK (CN) station number in B&R X20 system

Figure 35: Station number switches

Switch position Description

0x00 Reserved, switch position not permitted42

0x01 - 0xEF Node number of the POWERLINK station. Operation as a controlled node (CN).

0xF0 - 0xFF Reserved, switch position not permitted.

Table 17: X20BC0083 - Station number

NodeID 240 (0xF0) is a special case. This node number is reserved for the managing node (MN).
The range 1 to 239 (0x01 - 0xEF) is available for CNs.

Setting in Automation Studio

In Automation Studio, the NodeID is set as follows:
1) Right-click on bus controller shortcut menu and select
"Change node number".
2) Enter the new NodeID
3) The new NodeID appears in the "PLC Address" column
(red box)

Figure 36: Setting the NodeID in Automation Studio

41 Even personnel without special programming skills should be able to quickly and easily exchange a POWERLINK station.

42 Unless node supports dynamic node allocation (DNA)

44 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Automatic node number assignment

For stations that support it, the node number can also be assigned automatically during startup. This extended POW-
ERLINK feature is known as dynamic node allocation (DNA).
For modules with node number switches, DNA mode must be activated before it can be used. On B&R devices this is
done by setting NodeID = 0. A fixed NodeID must always be entered for the first station (from MN's perspective) in
a line ("Head of a line" node).

Communication / POWERLINK / General information / Dynamic node allocation (DNA)

IP addressing
If a POWERLINK node supports IP (Internet Protocol), then the last position in the IP address43 corresponds to the
node number:
■ 192.168.100.NodeID

A gateway is required in order to access this IP address from outside the POWERLINK network. The gateway remaps
the IP address using NAT (Network Address Translation).

Exercise: Incorrect node number setting

In this exercise we will be examining the effects of an incorrect node number setting using various diagnostic tools.

1) On CN1, the node number is set to "0x11"

2) Restart the CN (turn off/on) to apply the node number.

3) Observe the CN's LED status indicators.

4) Open System Diagnostics Manager (SDM). What clues can be gathered here (hardware view, diagnostic data
points, logger)?

5) Maybe open the logger for better readability, as well as the hardware comparison window in Automation Studio
(Online / Compare / Hardware).

6) Now set the node number of (formerly) CN1 to "2" and switch both CNs (formerly CN1 and CN2) off and back on.

7) Repeat the diagnostic steps and compare the results.

8) Perform a network capture with Wireshark and interpret the results of a startup phase.

The SDM overview provides an excellent starting point. Missing modules can quickly be identified in the
hardware view. Troubleshooting is particularly tricky if a node number is assigned twice in the POWER-
LINK network because of the irritating side-effects this causes. If you loose track, it may help to connect
devices to the network successively one at a time.

4.3 Diagnostics strategy

A systematic approach is absolutely essential for fast error diagnosis. To do this, you can gradually move from a su-
perficial approach for diagnosis to more detailed examination. Depending on the complexity of the error, appropriate
tools are used. The following procedure is recommended:
Step 1: Check the state of the CNs / verify successful startup
• Look at S/E and L/A LEDs
• Remote diagnostics with System Diagnostics Manager (SDM): ModuleOk" data point
• Remote diagnostics with Automation Studio: "ModuleOk" data point
43 In a POWERLINK segment the IP address always starts with 192.168.100.xxx.
POWERLINK – DIAGNOSTICS 45

Step 2: Investigation of failure to start up or sporadic failure

• Perform and evaluate network capture using
° Wireshark
° OmniPeek
° X20ET8819 for timing measurements and if error is not obvious

Step 3: Further investigation of faulty nodes

• Check network connections
° Check diagnostic data points in SDM or Automation Studio (e.g. "EthPhy1LinkOk")
° Read error counters (e.g. object 0x1C0B/0x01) using AsEPL library in Automation Studio
° RJ45 plug contacts
° Cable type
° Bend radius
° Cable drag chain suitability

• Check POWERLINK settings

° Timeouts
° Cycle time

• Minimum layout
° Reproduce error with minimum number of nodes
° Add CNs one at a time until error occurs

Most errors encountered in a POWERLINK network can be efficiently diagnosed using this method.
46 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

5 POWERLINK - Design and optimization

5.1 Network design and timing performance

Early in the process of designing a POWERLINK application, it is important to clearly define and verify the requirements
to be achieved by the network that connects the electronic components to one another. The more complex the network
and the higher the performance requirements of the application, the more essential this preparation becomes.

Figure 37: Typical topology of a POWERLINK network

5.1.1 Connection to other networks

The door to the "outside world": Gateways
In order to transfer data from a POWERLINK segment into other networks, such as an IP-based intranet or another
POWERLINK segment, you need to use a gateway or router (see Fig. 37 "Typical topology of a POWERLINK network"
on page 46).
Reasons a gateway is absolutely necessary:
■ Preserve the real-time capability of POWERLINK
■ Prevent overloading the 3rd party network with cyclic POWERLINK frames

A POWERLINK network is self-contained and is not permitted to be connected directly to another net-
work.

Routing and addressing mechanisms are used to interface between multiple IP-based networks. These are described
in Automation Help under:

Communication / Ethernet / Network settings

Communication / POWERLINK / FAQ

Basic Ethernet mode

A POWERLINK node can also function as a normal Ethernet node. When a POWERLINK CN does not detect any POWER-
LINK frames on the network, it switches automatically to BASIC_ETHERNET mode (see Fig. 19 "CN state machine" on
page 27). This is the only case where a POWERLINK node may be connected directly to a non-POWERLINK network. The
POWERLINK interface can be used as a normal Ethernet interface.
 POWERLINK - DESIGN AND OPTIMIZATION 47

5.1.2 Cycle time (MN parameters)

The cycle time is defined as the time between two SoC (Start of Cyclic) frames. The SoC is sent by the MN to mark the
beginning of a new POWERLINK cycle and synchronize all the stations in the network (see Fig. 4 "POWERLINK cycle"
on page 10).
The parameter "Cycle time" is included in the MN settings in the configuration tool (e.g. Automation Studio). The cor-
responding object number in the OD is 0x1006.
The shortest possible cycle time depends on the following parameters:
■ Number of nodes
■ Volume of cyclic data per node
■ Operating modes of individual nodes (3.2.2 "Isochronous phase" on page 11)
■ Maximum size44 of asynchronous frames (5.1.5 "Asynchronous MTU size (MN parameter)" on page 52)
■ Network latency (topology, hubs, cable length)

MN CN

PRes

■ Node response times (5.1.3 "PResMaxLatency", 5.1.4 "ASndMaxLatency")

These parameters can be used to calculate the cycle time. For additional information, see 5.1.2.1 "Esti-
mated cycle time requirements"

5.1.2.1 Estimated cycle time requirements

If you at least know the number and type of nodes in the POWERLINK network, you can quickly estimate the fastest
possible cycle time by using the following simplified assumptions:

Figure 38: Simplified formula for calculating the POWERLINK cycle time

IPG Interpacket gap (960 ns); mandatory pause time between two Ethernet frames (according to IEEE 802.3)

Hub level Number of hubs between MN and observed CN (not counting interface on CN itself)

The diagram of the frames in a POWERLINK cycle is a helpful reference when performing this calculation (see Fig. 4
"POWERLINK cycle" on page 10).

44 And number, in the case of Multi-ASnd

48 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Example calculation:
• Network layout:
1 CN receives 30 bytes, sends 136 bytes and is connected via an extra hub.
• Estimation of the shortest possible POWERLINK cycle time:
TCycle Min = 40 µs + 6 µs (PReq) + (6 µs + 8 µs) (PRes) + 3 µs (hub) = 63 µs

5.1.2.2 POWERLINK cycle time in Automation Studio

In Automation Studio the cycle time is set on the MN POWERLINK interface as shown in the following image:

The minimum cycle time value can be calculated for the topology in System Designer using the Hardware Configura-
tion Analyzer function:

Communication / POWERLINK / Calculations / Cycle time

More detailed information and explanations regarding the relationship between topology and timing can also be found
in Automation Help.

Communication / POWERLINK / Wiring

Settings in Automation Studio

Section 5.2 "Automation Studio settings for network timing" on page 53 describes where to find this parameter
in Automation Studio.

5.1.3 PollResponse timeout (CN parameter)

PollResponse timeout parameter
When a station is polled with a PReq frame, it takes a certain amount of time for it to respond with a PRes frame.
The PollResponse timeout defines the maximum length of time (in µs) that the MN will wait for a PRes frame from
a polled station (see Fig. 40 "Topology example: The MN checks the response delay"). This parameter should be at
least as high as the PResMaxLatency (see below), and must take the network topology (cable delay and hub level) into
consideration. The managing node waits to send the next (PReq) frame until the PollResponse timeout has triggered
on the managing node (see Fig. 41 "PollResponse timeout") or the PRes frame has been received.
In the configuration tool (e.g. Automation Studio), the "PollResponse timeout" parameter is included in the controlled
node settings. The corresponding object number in the managing node's OD is 0x1F9245.
Settings in Automation Studio
Section 5.2 "Automation Studio settings for network timing" on page 53 describes where to find this parameter
in Automation Studio.

45 Index: 0x1F92, Subindex: <NodeID of the CN>

POWERLINK - DESIGN AND OPTIMIZATION 49

Figure 39: PollResponse latency

Figure 40: Topology example: The MN checks the response delay

Figure 41: PollResponse timeout

50 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

This parameter depends on the topology of the network (cable lengths and hubs/switches increase la-
tency).

A line topology ("daisy chain") also includes an integrated hub for each controlled node (see Fig. 37 "Typ-
ical topology of a POWERLINK network" on page 46).

PollResponse timeout minimum: PResMaxLatency

The maximum PReq-PRes response time of a node (measured directly at the node) is called PResMaxLatency. The
PResMaxLatency value does not take topology delays or frame sizes into account and cannot be set in Automation
Studio, but is predefined by the properties of the node.
The PResMaxLatency value can be found in the controlled node's data sheet or device description file (see 6.1 "XDD -
Device description file"). The corresponding object number in the CN's OD is 0x1F98/0x03.

5.1.4 Asynchronous timeout (MN parameter)

Asynchronous timeout parameter
When a station receives permission from the MN to send asynchronous data, there is a certain delay between when
the SoA is sent and when the ASnd from the respective station is received. The Asynchronous Timeout parameter
defines the maximum length of time (in µs) that the MN will wait to receive a frame after sending the SoA – similar to
the PollResponse timeout (CN parameter) - see also Fig. 40 "Topology example: The MN checks the response delay".
The managing node waits to send the next asynchronous frame (for multi-ASnd) until the Asynchronous timeout has
triggered on the managing node (see Fig. 43 "Asynchronous timeout") or the asynchronous frame has been received.
The SoC frame is not delayed in case of a timeout.
Since this parameter is defined once for the entire network, it should be set to at least the maximum possible ASnd-
MaxLatency value (see below) of all stations.

Generally, the ASndMaxLatency value is the same as the PResMaxLatency value for a given station. In
this case, the asynchronous timeout can be set to the highest PRes timeout value in the network (CN
parameter in Automation Studio), since this value also accounts for topology-related delays46.

The parameter "Asynchronous timeout" is included in the MN settings in the configuration tool (e.g. Automation Stu-
dio). The corresponding object number in the MN's OD is 0x1F8A/0x02 and is called "AsyncSlotTimeout" in the POW-
ERLINK specification.
Settings in Automation Studio
Section 5.2 "Automation Studio settings for network timing" on page 53 describes where to find this parameter
in Automation Studio.

46 The difference in frame size between PRes and ASnd is not accounted for.
POWERLINK - DESIGN AND OPTIMIZATION 51

Figure 42: Asynchronous latency

Figure 43: Asynchronous timeout

This parameter depends on the topology of the network (cable lengths and hubs/switches increase la-
tency).

Minimum asynchronous timeout: ASndMaxLatency

The maximum SoA-ASnd response time for a node (measured directly on node) is referred to as ASndMaxLatency.
The ASndMaxLatency value does not account for topology-related delays or frame sizes and cannot be configured in
Automation Studio. It is predefined by the properties of the node.
The ASndMaxLatency value can be found in the controlled node's data sheet or device description file (see 6.1 "XDD -
Device description file"). The corresponding object number in the CN's OD is 0x1F98/0x06.
52 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

5.1.5 Asynchronous MTU size (MN parameter)

The asynchronous MTU47 size defines the maximum payload in bytes that an Ethernet frame in the POWERLINK asyn-
chronous phase is permitted to have.
The parameter "Asynchronous MTU size" is included in the MN settings in the configuration tool (e.g. Automation
Studio). The corresponding object number in the OD (of the MN or CN) is 0x1F98/0x08 and is called "AsyncMTU" in
the POWERLINK specification.

This value influences the shortest possible cycle time.

If a network is to be operated with short cycle times, this parameter must not be set too high. The lowest possible
setting is 300 bytes. If, on the other hand, a larger asynchronous throughput is more important for a particular appli-
cation, then the parameter should be set higher (max. 150048 bytes).

Figure 44: MTU size - Frame size affects cycle time

Settings in Automation Studio

Section 5.2 "Automation Studio settings for network timing" on page 53 describes where to find this parameter
in Automation Studio.

5.1.6 Loss of SoC tolerance (MN parameter)

CNs monitor the precision of the SoC frames sent by the MN. If an SoC is not received within a defined tolerance, it
is considered lost. The parameter "Loss of SoC tolerance" must therefore be set depending on the performance of
the MN.
The parameter "Loss of SOC tolerance" is included in the MN settings in the configuration tool (e.g. Automation Studio).
The corresponding object number in the CN's OD is 0x1C14.

47 Maximum Transmission Unit (Ethernet Payload)

48 Some IP stacks (on the CN) only support a standard Ethernet MTU of 1500 bytes. In this case, the value must be adjusted accordingly.
 POWERLINK - DESIGN AND OPTIMIZATION 53

This parameter should be set to a higher value (more than 2x the cycle time) if the POWERLINK MN is
running on a non-real-time system (e.g. openPOWERLINK under Windows XP). On a real-time capable
system like Linux/RTPREEMPT, it can be lowered to improve error monitoring performance.
With the B&R system, this value is not relevant due to the high-performance MN architecture and there-
fore cannot be set.

There is no way to modify this parameter in Automation Studio.

5.2 Automation Studio settings for network timing

The following images show where you can find the POWERLINK timing parameters in Automation Studio.
Further information can be found in Automation Help:

Select the parameter and press "F1".

Communication / POWERLINK / Calculations

MN parameters

Figure 45: MN parameters for network design in Automation Studio

54 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

CN parameters

Figure 46: CN parameters for network design in Automation Studio

Exercise: Verify the POWERLINK configuration

Use the Hardware Configuration Analyzer in Automation Studio to determine the POWERLINK network timing charac-
teristics. The configuration will automatically be verified in the process.

1) Change the "Response timeout" parameter of CN2 to 2 µs

2) Verify the POWERLINK parameters with <Open> / <Hardware Configuration Analyzer>

3) Fix the incorrect "Response timeout" value. The new value must be higher than the calculated minimum value.

Use the Hardware Configuration Analyzer to detect and resolve configuration errors before the project
is transfered to the target system. In order to do this, the actual network topology must be represented
in accurately in System Designer.

Diagnosis and service / Diagnostic tool / Hardware Configuration Analyzer

Exercise: Check the utilization of the POWERLINK cycle at runtime

Use Automation Studio to calculate the utilization of the POWERLINK cycle while the system is running.

1) Observe the "CycleIdleTime" data point on the POWERLINK MN interface (monitor mode).
POWERLINK - DESIGN AND OPTIMIZATION 55

2) Compare it with the configured cycle time on the POWERLINK MN.

3) Disconnect a random POWERLINK station from the network.

4) Observe the change in the "CycleIdleTime" data point.

The "CycleIdleTime" data point can be used to check the utilization of the POWERLINK network in SDM
or Automation Studio. This practical technique can be used in combination with the calculation of the
theoretical cycle time.

Communication / POWERLINK / Diagnostics / Diagnostic data points

Diagnostics and Service / I/O and network diagnostics / Recommendations for the application / Logging
diagnostic data points / POWERLINK data points

Exercise: Checking the POWERLINK cycle time

Use Wireshark to determine the actual cycle time.

1) Create a network trace of an active network (CN in OPERATIONAL state).

2) Calculate cycle time by right-clicking on an SoC frame / "Set Time Reference"

3) The "Time" column shows the time until the next SoC frame.

4) The format is set under "View / Time Display Format".

5) Change the POWERLINK cycle time in AS.

6) Compile the project and transfer it to the PLC

7) After the MN is restarted (automatically), check the cycle time with Wireshark.
56 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Wireshark can be used to measure timing. However, the displayed timestamps can only be relied on when
additional hardware support is used (X20ET8819). Be sure to account for network topology and hub level
in your measurements.

5.3 Mapping in Automation Studio

In Automation Studio, the mapping is further abstracted and is configured in the background. All one sees is a bus
station's cyclically transferred object, which can be linked to a process variable. This is not a direct link – it requires
a number of intermediate steps.
MN to CN connection
When a POWERLINK node is added to an Automation Stu-
dio project, its cyclic data points are available in the I/O
mapping49. For non-B&R devices, the channel name in the
I/O mapping also includes the object number.

Figure 47: I/O mapping of a POWERLINK module

With POWERLINK, an I/O mapping channel corresponds to a specific (cyclically) transferred object that
exists on the selected node. All activated I/O mapping channels for CNs are summarized in the "process
image" on the MN.
The two process images (input and output process image) serve as the interface to the PLC application.

The cyclic objects of a POWERLINK node never actually exist in the object dictionary of another node (e.g. the MN), they
are simply represented in its local memory (see Fig. 13 "The content of mapping objects defines correlation between
network and application" on page 20).
Automation Studio / Automation Runtime handle all of the configurations required for an OPERATIONAL CN to transfer
cyclic objects to the MN application.
Configuration set in the background by the MN:
■ Connection of frame data to object: CN mapping
■ Connection of frame data to process image: MN mapping
■ Enable programming access (MN): Connection of process variable to process image

For additional information about the POWERLINK node configuration in Automation Studio, see the help documenta-
tion. The explanations in sections below apply to both 3rd-party devices and B&R devices.

Communication / POWERLINK / 3rd-party devices / Configuration of OEM devices

CN to CN - Direct cross-communication
In Automation Studio, a cross-communication link (CN to CN) can be established with little effort. The CNs are con-
figured via Automation Runtime. The procedure presented in the help documentation applies to other nodes besides
the described iCNs:

Communication / POWERLINK / Intelligent Controlled Node / Cross-communication between iCNs

In a data link with cross-communication, an (RPDO) frame error – caused by electromagnetic interference,
for example – can only be detected by the application. Such an application might take the form of a
cyclically incremented counter that is also contained in the PDO and is monitored by the receiving node.
With a normal PReq-PRes data link, a (repeated) frame error would cause a state change in the MN or CN.

49 For CNs that support dynamic mapping, a cyclic channel must first be activated in the I/O configuration.
 POWERLINK - DESIGN AND OPTIMIZATION 57

5.4 Optimization options in Automation Studio

A POWERLINK offers numerous parameters that can influence the performance of the network. This section will intro-
duce and describe the most important parameters.

5.4.1 Cycle time and possible number of nodes

The following Automation Studio settings can be used to shorten the POWERLINK cycle time or increase the maximum
number of nodes (CNs participating in the full cycle) in a POWERLINK network.

5.4.1.1 Multiplexing
Multiplexing prescaler (MN parameters)
This parameter defines the prescaler for multiplexing stations. It determines the number of POWERLINK cycles across
which the multiplexed nodes should be distributed. A value of 3 means that a multiplexing CN is addressed in every
third cycle (see "Multiplexing" on page 13). The number of multiplexed stations is divided by the multiplexing prescaler,
and the result is rounded up to the next whole number. This results in the number of multiplexed stations per cycle.

If you multiply the multiplexing prescaler by the POWERLINK cycle time, the result is the "cycle time of a multiplexing
node", i.e. the time it takes to update the data of a node for which multiplexing is activated.
This parameter is only available if the MN supports multiplexing.
Multiplexed station (CN parameters)
Multiplexed mode is activated for each CN individually as shown below.
Automation Studio automatically calculates the multiplexing slot in which a CN is addressed. The slot can also be
defined manually under the "Advanced multiplexing" settings.

This parameter is only available if the CN and MN both support this mode.
58 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Communication / POWERLINK / Multiplexed

5.4.1.2Poll response chaining

Chained station (CN parameters)
A CN can be operated in poll response chaining mode (see "Poll response chaining" on page 12) by activating the fol-
lowing parameter:

Poll response chaining is activated for each CN individually as shown below. This parameter is only available if the CN
and MN both support this mode.

For additional information about PRes chaining, see the POWERLINK specification and Automation Help.

Communication / POWERLINK / General information / Poll response chaining

5.4.1.3Direct cross-traffic

Another way to reduce the required POWERLINK cycle time is using direct cross-traffic (see "Direct cross-traffic" on
page 12) for CN-to-CN communication. An example of an application using direct cross-traffic is for motor control in
cases where a drive needs to read the position of a master axis to synchronize movement of coupled axes.

Programming / I/O configuration / I/O mapping / Cross-traffic

Programming / Editors / Configuration editors / I/O mapping / Select variable dialog box / Channels
Communication / POWERLINK / General information / Cyclic data exchange

CN to CN communication via the MN

Without direct cross-traffic, network stations can only communicate by way of the MN, which copies the data and
forwards it on to the other CN. The image below illustrates the steps involved in this process for an Automation Studio
project.
POWERLINK - DESIGN AND OPTIMIZATION 59

Figure 48: CN to CN connection via the MN

void _CYCLIC CopyAItoAOCyclic(void)

{
Program code
AnalogOutput = AnalogInput;
}

Table 18: ANSI C program "CopyAItoAO"

Data path:
1) The first CN (bus controller with AI module) sends analog input data to the MN in the PRes frame.
2) The MN receives the data and makes it available to its application in a process image.
3) The process image is mapped to a process variable.
4) The contents of the process variable for the analog input module (AI) are copied to the process variable for the
analog output module (AO) in a task.
5) The AO process variable is mapped to the process image for the AO module.
6) The MN sends the AO data to the second CN (bus controller with AO module) in the PReq frame.

Direct CN to CN communication
This data path via the MN can be eliminated by having the CNs communicate with each other directly.
60 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Figure 49: Direct CN to CN connection

With the right configuration, a PRes frame can be received by any other CN in the network. This configuration of a
"monitoring" node generally takes place during startup, but can also occur later while the node is in the OPERATIONAL
state. The path taken by the data is reduced to a single step, as shown in the image above (Fig. 49 "Direct CN to CN
connection").
Data path:
1) The first CN (bus controller with AI module) sends analog input data to the MN in the PRes frame, and the second
CN (bus controller with AO module) receives the data.

The mapping of a "monitoring" CN needs to be configured (on the MN) so that it successfully links the
received PRes data to the AO module. A cyclic data connection to the MN does not exist here, however.

Exercise: Examine the data path for CN-to-CN communication.

In this exercise, we will send a CN's input data first to the MN, where it will be copied and forwarded on to the second
CN, as illustrated in Fig. 48 "CN to CN connection via the MN". Then we will compare this to the path for direct CN-to-
CN communication ( see Fig. 49 "Direct CN to CN connection").
 POWERLINK - DESIGN AND OPTIMIZATION 61

Figure 50: Panel elements and their data points in the AS sample project "POWERLINK_ETA"

Overview of panel functions with Automation Studio sample project "POWERLINK_ETA":

■ Potentiometers P1 and P2 on the control panel are mapped to the first two inputs of the X20AI4622 module.
■ The two 7-segment displays are mapped to the first two analog outputs of the X20AO4622 module.
■ The analog values of Potentiometer P1 and P2 appear on the left and right 7-segment displays, respectively.
■ Data is exchanged between potentiometer and display via POWERLINK, following a different data path for each
display.

Table 19: Description of POWERLINK ETA test system functions

1) Start Automation Studio project "POWERLINK_ETA" and transfer the project to the PLC50

2) Establish online connection with the PLC.

3) Start monitor mode

4) Force data point AnalogOutput02 in the I/O mapping of the X20AO4622 module using any values

5) Observe the seven-segment display

6) Compare with your own project and examine the path taken by the analog data using Fig. 48 "CN to CN connec-
tion via the MN" on page 59 as a reference.

7) Force data point AnalogOutput01 in the I/O mapping of the X20AO4622 module. Why is this not possible?
Note: The process variable entered in the I/O mapping is AnalogInput01 of module X20AI4622, but is not a PLC
variable like in the previous case. Compare with Fig. 49 "Direct CN to CN connection" on page 60 .

50 . Program CompactFlash card

62 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

8) Change the I/O mapping so that the lower seven-segment display also receives the data from potentiometer P1,
which is connected to module X20AI4622. Here, direct cross-traffic is to be used as the connection type:

9) Perform a network capture and save it as Direkt_Crosslink.pcap (for later use).

The POWERLINK connection between analog input 2 and analog output 2 goes through the MN and can
therefore be influenced by the MN. The (already digitalized) data take the following path:
■ X20AI4622 / X2X / Bus controller A / PRes / MN / PLC task / MN: PReq / Bus controller B / X2X /
X20AO4622

In contrast, the direct connection (with direct cross-traffic) between AI1 and AO1 cannot be influenced
by the MN since the data never passes through it. The data takes the following path:
■ X20AI4622 / X2X / Bus controller A / PRes / Bus controller B / X2X / X20AO4622

Exercise: Interrupt the data path on the MN.

With a data link between two CNs via the MN, interrupt the connection on the MN. This demonstrates the effects on
the receiving station and the functionality of cyclic data exchange.

1) Change the connection between AnalogInput02 and AnalogOutput02 in the Automation Studio project "POWER-
LINK_ETA" back to a CN-MN-CN path (no direct cross-traffic, see Fig. 48 "CN to CN connection via the MN").

2) Compile the project and transfer it to the PLC

3) Interrupt the CN-MN-CN data path by stopping the copy process on the MN (PLC). To do this, open the Watch
window for the object "CopyAItoAO" and stop cyclic execution.

4) Turn the potentiometer "P2"

The MN periodically sends the latest output data value to CN2. From POWERLINK's point of view, data
transfer works fine, but the output data is no longer updated by the PLC. In the Watch window you can
see that only the input data (AnalogInput) is being updated.

Exercise: Compare data paths and the effects of direct cross-traffic.

In order to complete this exercise, you must first work through section 4.1.5 "Wireshark".

Trace the path of cyclic data (PDOs) between two CNs in Wireshark. Compare two different approaches, each involving
different POWERLINK frames in the data link. Also observe the effects on the shortest possible cycle time.
 POWERLINK - DESIGN AND OPTIMIZATION 63

1) Perform a network capture (using Wireshark) and save it as Link_over_MN.pcap.

2) Compare the two captures Direkt_Crosslink.pcap (from previous exercise) and Link_over_MN.pcap.

■ Make a note of the differences

■ Calculate the theoretically required cycle time (using auxiliary formula in section 5.1.2.1 "Estimated cycle time
requirements" on page 47) for both network recordings

When there is direct cross-communication between two CNs without a "detour" through the MN, no data
is sent in the PReq frame to the "output data bus controller", since it gets all the data it needs straight
from the PRes frame of the "input data bus controller". Only the minimum size of the PReq frame needs
to be sent to the bus controller, since it contains no data.
Direct cross-traffic can therefore have a very positive effect on the overall system:
■ Node-to-node response time decreases
■ Theoretically required cycle time decreases
■ Increases max. number of CNs in the POWERLINK network
■ Reduces the load on the MN because it does not have to copy the data

The best use of direct cross-traffic is obtained in combination with 5.4.1.2 "Poll response chaining" or
with several cross-traffic connections with poll response chaining for the transmitting and Multiplexing
for the receiving nodes.

5.4.2 Asynchronous bandwidth - Multiple asynchronous send

For a more detailed explanation of this POWERLINK extension, see section 3.2.3 "Asynchronous phase" on page 14.
Asynchronous slots per cycle (MN parameter)
This parameter can be used to increase the number of asynchronous frames that can be sent per cycle.

An additional asynchronous frame in the POWERLINK cycle can only be transmitted by a node that sup-
ports Multiple ASnd. The MN must also support the feature of course.

For more information, see the Automation Help.

Communication / POWERLINK / Multiple asynchronous send

64 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

5.4.3 Response time of the overall system

A PLC system normally includes three basic components:

■ Operating system (Automation Runtime)
■ Backplane bus / internal data bus (X2X)
■ Fieldbus (POWERLINK)

The way these components interact can be optimized for the needs of a particular application using various settings
in Automation Studio.

The following image shows the numbered individual steps that add up to the overall system response time (total
latency). For additional information, see Automation Help.

Communication / POWERLINK / Response time

For iCN: Communication / POWERLINK / Intelligent Controlled Node / Response time

Figure 51: Response time of the overall system

The overall system response time is comprised of:

■ Input latency (1-5)
■ Data processing / task duration (6)
■ Output latency (7-12)

Latency optimization
Automation Studio offers a number of ways to reduce the overall response time of a PLC system. These configuration
parameters and the effect they have on the system are described in Automation Help (see previous help documentation
link). The direction of data transfer (e.g. for input/output latency) is specified from the perspective of the PLC.

For these settings to have an effect, it is important that all cyclic processes are synchronized, i.e. that all
cycle times are whole number factors or multiples of each other.
Automation Runtime system timer must also be synchronized with the POWERLINK cycle.

Most of these configuration parameters are only available for B&R devices and are not included in the POWERLINK
specification, since their influence extends beyond the POWERLINK network.
Synchronization of the overall system
For a detailed description of synchronization between the individual system components and their hardware depen-
dencies, see Automation Help under:

Communication / POWERLINK / Synchronization

With SG4 target systems, for example, task classes are synchronized with the POWERLINK network using the settings
under <CPU> → Settings → Timing → System timer.
POWERLINK - DESIGN AND OPTIMIZATION 65

Figure 52: Synchronizing tasks with the POWERLINK network in SG4 systems

Optimization limits
Optimizing latency always comes at the cost of available CPU processing time and the size of frames that can be
transferred in a single bus cycle. In other words, the latency can only be optimized if a task is not very time-intensive
and the volume of data to be transferred in each cycle is not to large.
Where the limit lies here depends strongly on the application and can therefore only be determined in practical situ-
ations.

Generally, you have to choose between an optimal response time and stable synchronization of stations
with the POWERLINK cycle.
 66 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

6 POWERLINK - Third-party device sup-

port
6.1 XDD - Device description file

Every device with a POWERLINK interface must include a description of its char-
acteristics in the form of an XML device description (XDD) file.
This file can be interpreted by a configuration tool in order to configure the indi-
vidual nodes and in turn the entire POWERLINK network (from the MN).

The XDD file describes the object dictionary and other network and device parameters of the respective
POWERLINK node.

An XDC file is an expanded form of the XDD file.

This file contains deviating configuration values in addition to the default values of an object; otherwise,
there are no differences.

The XDD file is comparable to the configuration files for other bus systems:
■ GSD: PROFIBUS
■ EDS: (CANopen, DeviceNet)

The format is standardized. Sample files and additional information are available on the EPSG website (www.ether-
net-powerlink.org) and in Automation Help:

Communication / POWERLINK / 3rd-party devices

XDD checker
It is possible to have an XDD file checked for errors. You can upload a file to the EPSG website and receive an email
listing any errors that are found.
■ www.ethernet-powerlink.org / Service

Exercise: Add a rotary encoder to an Automation Studio project.

Add a 3rd-party device to an Automation Studio project and integrate it into the POWERLINK network.

1) Download the XDD file from the manufacturer's website51.

2) Connect the rotary encoder to the PLC and power supply.

3) Import the XDD file. See section of Automation Help referenced above. Here you will find a description of how to
import an XDD file into Automation Studio.

51 Alternatively, your trainer may provide you with a file.

 POWERLINK - THIRD-PARTY DEVICE SUPPORT 67

4) Update the PLC configuration

5) Check the POWERLINK status of the rotary encoder using the S/E LED.

6) Rotate the encoder shaft and observe changes to the input data (monitor mode)

Automation Studio's XDD import feature makes it easy to integrate 3rd-party devices in the POWERLINK
network. The XDD file must be obtained from the manufacturer of the device.

Exercise: Questions regarding integration of a rotary encoder in a POWERLINK network

1) Why does not a cyclic data channel need to be activated in Automation Studio for the rotary encoder?

2) Why are the mapping objects (see Tab. 6 "Object range for PDO mapping objects." on page 19) not written during
startup (perform Wireshark trace)?

3) What does the write access to object 0x1010/0x01 with the content "Save" mean52 (according to the POWERLINK
specification)?

4) What does the write access to object 0x1011/0x01 with the content "Load" mean (according to the POWERLINK
specification)?

6.2 Device description file for modular devices

If a device is made of several different and replaceable modules that transfer their data via POWERLINK, it is possible to
describe each module with a separate file. The configuration tool imports more than just an XDD alone; it also imports
modular XDD files of all "child modules" as well as the "head station" as can be seen in Fig. 53 .
Modular XDDs for the X20 system can be found here:
■ www.br-automation.com / Downloads / Control and I/O systems / X20 system / Bus controller / X20BC0083 /
Fieldbus device description files

Figure 53: Example for the description of modular devices with XDD files

52 In order to observe access to 0x1010 in the Wireshark trace, you may first have to use the AsEPL library in Automation Studio to write the object 0x1011/0x01 with the content
"load" and then turn the node off/on. This can happen as described in "Exercise: Read objects using the SDO protocol." on page 32, with SDO write access instead of SDO read
access.
68 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

Modular XDD files are used to import modular 3rd-party devices to any configuration tool. The X20 system
can be used as a CN with an openPOWERLINK-based MN, for example.

6.3 XDD application example: openPOWERLINK MN

As can be seen in the following image, individual XDD files of various CNs can be imported to a configuration tool
whereby an overall network configuration is created. In the example with an openPOWERLINK MN, the configuration
tool is called "openConfigurator". If a B&R MN were used, then the configuration tool would be Automation Studio.
The configuration is provided to the openPOWERLINK MN as a CDC file in order to write the CNs' parameters (objects)
using the SDO protocol and configure the network.

Figure 54: XDD application example: Configuring a POWERLINK network with an openPOWERLINK MN

6.4 Examples of non-B&R MNs:

6.4.1 openPOWERLINK

openPOWERLINK is an open source implementation of the POWERLINK protocol that can be used either as a managing
node (MN) or a controlled node (CN). This implementation of the POWERLINK protocol is also called a "protocol stack"
or simply "stack". The openPOWERLINK stack was developed as a generic implementation that can easily be ported to
various hardware and software platforms. The Intel PC architecture (x86) under GNU/Linux is used as the reference
platform for openPOWERLINK. Ported versions are available for Windows, VxWorks and FPGA platforms, among others.
openPOWERLINK can be downloaded here:
■ http://sourceforge.net/projects/openpowerlink/

6.4.2 CODESYS

A POWERLINK MN can also be implemented using the CODESYS tool chain and specific PLC manufacturers.

■ http://www.be-services.net / Products / POWERLINK integration in CODESYS

 SUMMARY 69

7 Summary
Although Automation Studio makes working with the POWERLINK protocol exceptionally easy, the full benefit of this
powerful technology can only be enjoyed with a more comprehensive understanding of its inner workings. That was
the motivation for creating this training module. The better you understand the processes involved, the faster you
will be able to set up and optimize a POWERLINK network, and the more effectively you will be able use the available
diagnostic tools.
When you begin troubleshooting, it is best to try the most basic diagnostic features first. These include the LEDs on
each POWERLINK node as well as the messages and data points that are provided in Automation Studio or System
Diagnostics Manager.

If these approaches are unsuccessful, the next course of action is a more detailed network capture. Further analysis
is then generally conducted by the device manufacturer on the basis of this capture. It is important for the actual
network behavior to be reflected with precise timing. This ensures that further analysis based on the capture will lead
to successful results.
70 POWERLINK CONFIGURATION AND DIAGNOSTICS TM950

8 Appendix – Solutions to the exercises

"Exercise: Mapping questions" on page 24
1) 0x1400 to 0x1AFF: Object area for PDO mapping objects
2) A pair of objects that always exist together, i.e. 2 objects: "Communication and mapping parameters"
3) Each node has its own object dictionary. Only 2 nodes are involved in direct cross-communication, so the answer
is: 2. With a CN-MN-CN connection, 3 nodes are involved – so 3 object dictionaries.
4) Not at all. The mapping only describes cyclic data transmission with PDOs.
5) The link between the network (PDO frame) and the source/target object in the application. (But not: The connec-
tion between two nodes!)
6) With the first two numbers of the mapping parameter: 0x18xx corresponds to send direction, 0x16xx receive di-
rection.
7) Objects 0x1400 and 0x1600 are already reserved for the PDO mapping of the PReq frame, since each CN must be
addressed by a PReq frame (or PRes MN frame in the case of PRes chaining).
8) Length: 48 bits, Offset: 32 bits Subindex: 0xFF, Index: 0x4000

"Exercise: Questions regarding integration of a rotary encoder in a POWERLINK network " on page 67
1) The rotary encoder uses a preconfigured mapping that cannot be modified, also called a "static mapping". A data
channel must only be activated in the I/O configuration if the mapping of the node can be configured dynamical-
ly, like the bus controller for example.
2) Reason: Also static mapping. The mapping is not modified; it is already provided.
3) See POWERLINK specification "EPSG DS 301 V1.x.x - Communication Profile Specification": Save the current val-
ues of all parameters that can be saved in nonvolatile memory for later use. This means these parameters will not
have to be reconfigured via SDO and instead can be loaded automatically at the next startup.
4) See POWERLINK specification "EPSG DS 301 V1.x.x - Communication Profile Specification": Load the default con-
figuration for parameters that can be saved in nonvolatile memory. This resets the node configuration. The de-
fault configuration is not activated until the node is restarted.
 AUTOMATION ACADEMY 71

Automation Academy
Your knowledge advantage
The Automation Academy provides targeted training courses for our customers as well as for our own employees.
Expand your skills in the field of automation technology and learn to independently implement efficient automation
solutions using B&R systems.

Decide for yourself which learning concept you prefer!

Classroom learning
B&R offers standard seminars at all B&R locations. Services include seminar documents, effective communication of
training course content by experienced trainers and an Automation Diploma. A combination of group work and self-
study provides the high level of flexibility needed to maximize the learning experience.

Virtual classroom
Remote Lectures supplement B&R's continuing education portfolio with a virtual classroom, offering an alternative to
our on-site seminars. Selected content from our standard seminars is offered online. In addition to remote learning
methods, powerful simulation tools and secure remote maintenance are used.

Online courses
Take control of the content and learn at your own pace. With B&R online courses, you can take your first steps in the
world of B&R automation at any time. Based on a comprehensive narrative, you will independently work out how to use
our products. The mix of different media allows a logical sequence to be followed when learning as well as a targeted
choice of information to be used as a reference.

Contact
Would you like additional training? Are you interested in finding out what the B&R Automation Academy has to offer?
If so, this is the right place.
Access additional information here:
https://www.br-automation.com/de/academy/

Enjoy your next training course!

B&R
Industrial Automation GmbH
A member of the ABB Group
B&R Straße 1
5142 Eggelsberg, Austria
[email protected]

t +43 7748 6586-0

f +43 7748 6586-26

br-automation.com

V2.0.0.0 ©2023/10/30 by B&R, All rights reserved.

All registered trademarks are the property of their respective owners.
Subject to technical changes without notice.