Ethernet Powerlink
Ethernet Powerlink (official spelling: Ethernet POWERLINK ) is a real-time Ethernet to real-time data to be transmitted in the microsecond range. The main purpose is the transmission of process data in automation technology .
Ethernet Powerlink was originally developed by B&R Industrial Automation and is now being further developed and specified as an open standard by the open user and provider group EPSG (Ethernet Powerlink Standardization Group).
Overview
Developed from the outset with a focus on standard conformity, Ethernet Powerlink introduces a mixed polling and time slice mechanism for the deterministic transmission of data. This achieves:
- the guaranteed transmission of time-critical data in very short isochronous cycles with configurable time behavior
- the time synchronization of all network nodes with very high precision in the sub-microsecond range
- the transmission of the less time-critical data volume in the reserved asynchronous channel
Current implementations of Ethernet Powerlink achieve cycle times of less than 200 µs and a temporal precision ( jitter ) of less than 1 µs.
Ethernet Powerlink also specifies a communication protocol based on CANopen for the exchange of user data with nodes in the network. Both parts together are handled by a Powerlink protocol stack. No special hardware is required for this, so that both master and slave nodes can be implemented with commercially available Ethernet modules. Open source master and slave stacks are therefore also available for different operating systems.
Data transfer
Since Ethernet Powerlink is located on layers 2 and 7 in the OSI layer model , it is basically independent of the physics used (layer 1). In practice, however, it is mainly operated with twisted pair cables as Fast Ethernet (100Base-TX). Both the commercially available 8P8C / RJ-45 and the industrial M12 connectors are permitted. The use of fiber optic cables is also possible, although the additional delays caused by media converters must be taken into account.
For clean cabling, the Ethernet Powerlink standard refers to the IOANA guidelines for planning and installing industrial networks (IAONA's Industrial Ethernet Planning and Installation Guide ). In order to minimize the delay and jitter, repeating hubs are recommended instead of switching hubs ( switches ) within the real-time domain .
Versions
Ethernet Powerlink currently exists in two forms:
Version 1 (Ethernet type 0x3e3f) is a proprietary approach from B&R, which, however, was opened early as a basis for further developments.
Version 2 (Ethernet type 0x88ab), on the other hand, is the current, published standard of the EPSG, which has been expanded to include various mechanisms (CANopen device profiles, Powerlink Safety, Electronic Datasheets, Master Poll Response).
Although both types of protocol are very similar, the following information relates only to version 2, since version 1 is only intended as a temporary solution. Some participants can work with both protocol variants.
In 2006 the EPSG announced the development of Gigabit Powerlink.
Data format
Each Powerlink package consists of a header and the actual user data. This packet is integrated in a normal Ethernet frame, which must have a size between 64 bytes and 1500 bytes. Jumbo frames (> 1500 bytes) are not allowed in a Powerlink network. As EtherType for Ethernet Powerlink was 0x88AB assigned by the IEEE.
The Powerlinkheader itself consists of:
- 1 bit reserved
- 7 bit MessageType
- 8 bit destination node number
- 8 bit source node number
The following message types are defined:
| MessageType | ID | Name of the message | use | Ethernet transfer type |
|---|---|---|---|---|
| SoC | 0x01 | Start of cycle | Defines the start of a new cycle | Multicast |
| PReq | 0x03 | PollRequest | Request cyclical data from the CN | Unicast |
| PRes | 0x04 | PollResponse | Send current cyclic data of the CN | Multicast |
| SoA | 0x05 | Start of asynchronous | Signal the start of the asynchronous phase | Multicast |
| ASend | 0x06 | Asynchronous send | Sending asynchronous data | Multicast |
Real-time communication
In order to guarantee deterministic data transmission, collisions on the network must be avoided. For this purpose, the data transmission is controlled by a special participant, the Managing Node (MN) . The individual network participants, the Controlled Nodes (CN) are only allowed to send if they have been specifically requested to do so.
A cycle begins with the message Start of Cycle (SoC) . Each node is then queried individually by the MN with a poll request (PReq) , to which the CN replies with a poll response (PRes) . Since the responses are sent as an Ethernet multicast, other Powerlink devices can listen in. Cross traffic between the CNs is thus possible. In order to keep the cycle time small, not every device has to be queried in every cycle ( multiplexed stations ). The response time of a device ( ) is an important quality feature.
After completion of the cyclical phase, the asynchronous phase begins with the Start of Asynchronous (SoA) package. In this phase, a CN determined by the MN can send non-cyclical data. In the asynchronous phase, data from a normal, non-deterministic network and the Powerlink network can be exchanged via special gateways.
Object directory
Based on the CANopen fieldbus standard , all communication objects and all user objects are combined in an object directory (OD) with Ethernet Powerlink. In the Powerlink device model, the object directory is the link between the application and the communication unit. Each entry in the object directory stands for an object and is identified by a 16-bit index. Each index can in turn contain up to 256 sub-indices. This means that up to 65536 × 254 user entries can be distinguished per device. (The sub-indices 0 and 255 cannot be used freely.) In profiles, the assignment of communication and device profile objects to a respective index is precisely defined, and thus the object directory defines a unique interface between the application and external communication.
| Index area | use |
|---|---|
| 0000 | not used |
| 0001-009F | Data types (special case) |
| 00A0-0FFF | reserved |
| 1000-1FFF | Communication profile |
| 2000-5FFF | manufacturer-specific area |
| 6000-9FFF | up to eight standardized device profiles |
| A000-BFFF | standardized interface profiles |
| C000-FFFF | reserved |
Device profiles
Device profiles have been defined for a number of device classes. These device profiles define the functionality and structure of the object directory for the respective devices. By using devices that correspond to a certain profile, greater independence from device manufacturers is achieved. Ethernet Powerlink uses the device profiles of CANopen . Transformation rules determine which objects of the CANopen device profiles are used for Powerlink devices. This takes into account the fact that the length of the user data packets (PDOs) is greater with Powerlink.
Electronic datasheets
Electronic data sheets are required to use Powerlink devices. These are stored as XDD files (XML Device Description), which correspond to the standardized XML format according to ISO 15745-4, and describe both the most important parameters of the objects in the object directory of a device and other parameters such as: B. the supported communication services. Configuration tools can read data sheet files and use them to communicate with the respective device and, if necessary, parameterize it.
standardization
Ethernet Powerlink was in the standards IEC 61784 -2, IEC 61158 was added -3, IEC 61158-4, IEC 61158-5 and IEC 61158-6. (The IEC 61784-2 standard specifies communication profiles, and the IEC 61158 standard specifies services and protocols for fieldbuses.)
Any Ethernet frames can be sent in the asynchronous phase. Therefore u. a. all IP -based protocols on higher layers, such as TCP , UDP and above, are used in the Ethernet Powerlink network. In detail, Ethernet Powerlink supports the following standards:
- IEEE 802.3 (Fast Ethernet)
- IP-based protocols (ICMP, UDP, TCP, ...)
- Standard device profiles: CANopen EN 50325-4 for automation
- IEC 61588 for real-time domain synchronization (future versions)
diagnosis
Standard diagnostic tools such as Wireshark ( free software ) or Omnipeek (commercial) can be used. Corresponding tools are listed on the EPSG homepage for more targeted diagnosis.
Transmission of safety-critical data
For safety-critical applications, Powerlink can be expanded with the additional, open safety protocol openSAFETY (formerly Ethernet Powerlink Safety). With openSAFETY, the safety-critical data is divided into two subframes and secured with checksums. The network's safety function is provided by its own safety controller. Safe and non-safe participants can coexist in a network and also exchange data that is not essential for the safety function.
openSAFETY is implemented as a protocol for the application layer. As such, it can be implemented on a variety of Industrial Ethernet network topologies. openSAFETY has been tested by TÜV Rheinland and TÜV Süd and approved for use in safety-critical applications in accordance with IEC 61508 SIL 3 and Category 4 of the Euronorm 954-1.
Others
Ethernet Powerlink should not be confused with Power over Ethernet , the power supply via the (unused) wire pairs, or with PowerLAN .
Web links
- http://sourceforge.net/projects/openpowerlink - Open source Powerlink stack for LINUX, Windows and FPGAs
- http://sourceforge.net/projects/openconf - Open source configuration program for Powerlink
Individual evidence
- ↑ Industrial Ethernet Planning and Installation Guide ( English , PDF) IAONA. Retrieved April 1, 2019.
- ↑ EPSG
- ↑ openSafety. Uniform standard for secure networks , SPS-Magazin , ETH3 2010 (from August 12, 2010).