DeviceNet Overview

TURCK 6/5/2018
What is it?
DeviceNet is a Low-Level Fieldbus Network that eliminates hard wiring and connects industrial devices to higher level programmable controllers.
What are its basic components?
Scanner, DeviceNet cables and cordsets, I/O nodes, terminating resistors, power supply and grounding wire.
Where is it used?
Automation, safety devices and large I/O control networks.
Who is responsible for it?
DeviceNet was originally developed by American company Allen-Bradley (now owned by Rockwell Automation).
The DeviceNet is a low-level fieldbus network that eliminates hard wiring and connects industrial devices such as limit switches, photoelectric sensors, valve manifolds, motor starters, process sensors, bar code readers, variable frequency drives, panel displays and operator interfaces to higher level programmable controllers. It supports 64 nodes, attached to the network in trunkline/dropline topology, as shown on Figure 1. The trunk is the main communication cable that distributes 24VDC power and communication data to all nodes through the DeviceNet media: cables, cordsets, tees, multiport junction boxes, power taps and terminating resistors.

The length of the trunk depends on the data rate and type of cable. The common practice is to use a thick cable for the trunk, whose length is limited to 1,640ft (500 meters). A thin cable is used for the drops. The length of the drop is limited to 20 ft. (6 meter). It is measured from the trunk to the farthest node at the drop. The ends of the trunk are terminated using two 121 Ohm terminating resistors. The network must be grounded at a single location only. Multiple power supply units are allowed but only one of them is grounded. The best location to ground the DeviceNet is in the middle of the network.
The DeviceNet is a connection-based network. There are two types of connections, the Explicit connections (point-to point or peer-to-peer) and IO connections. The Explicit messages are used for the network configuration, node commissioning and IO connection initialization. They are determined by a service code (command) and destination designators: Class, Instance and Attribute. Once the   command is executed, the connection is closed.  The IO connections (Bit-strobe, Poll, Change-of-state/COS and Cyclic) are used for the continuous IO data exchange between a scanner and nodes. All messages are organized in 4 groups, where group 1 has the highest priority on the bus and group 4 the lowest. Each DeviceNet device has integrated a CAN controller which is used for communication.
It generates noise immune, differential, communication signals that carries data over the network. CAN uses a bitwise arbitration method called CSMA/BA (carrier sense multiple access / bitwise arbitration), that assures the highest priority message always gets access to the bus in the event of  multiple device requests for data transmission. The CAN stands for the Controller Area Network as defined by the Bosch CAN Specification V2.0 and ISO 118981 standard.  These standards are the foundation of the Common Industrial Protocol (CIP), the DeviceNet adaptation of CIP (the DeviceNet Specification), by Open DeviceNet Vendor Association, Inc. (ODVA).
Interacting With DeviceNet
There are three levels of interaction involving DeviceNet applications:
Gateway (G/G) makes data exchange between DeviceNet and other fieldbus networks possible. They are used as alternative or supplemental devices to reduce the load on the standard networking techniques or to simplify interconnections. A few examples are: DeviceNet/EtherNet IP, DeviceNet/PROFINET and DeviceNet/ASi. A gateway usually appears as a single node on a higher level network and as a scanner on a lower level network. ASi gateway appears as a single node on the DeviceNet and a master on the ASi network.
Bridge (B/B) provides simple and the least expensive way to exchange data between two DeviceNet networks. The bridge, also called a Spanner, consists of two DeviceNet nodes which are electrically and optically isolated from each other. Each node is configured with a scanner where is resides. Each node has individual address switches and Autobaud capability. The data size exchanged between bridge nodes is flexible and may be selected during node configuration.
Repeater (R/R) or bus extender is used to extend the length of the network drop beyond 20ft limitation. The Repeater may be used to create Y shape network or in a warehouse facility multiple extended drops each up to 1640ft long at 125kB. The DeviceNet design rules apply to the extended network segment (B) the same way as they apply to the main network segment (A). The extended segment must have a separate DeviceNet power supply unit; it must be terminated at both ends and grounded at a single point. Each network segment may have multiple nodes, whose addresses are unique for the entire network. All nodes are set to the same data rate. The total number of nodes on all network segments cannot exceed 64 nodes. The number of the repeaters is not limited; do not cascade repeaters as each one introduces 2ms transmission delay.
Basic Parts List
A typical system consists of the following parts:
•  Master
•  Power Supply
•  DeviceNet I/O Modules (slaves)
•  Power Cables
•  Network Cables
System Configuration
DeviceNet stations require a network master (also called a scanner) to interface the stations to the host controller. TURCK DeviceNet stations are designed to be fully compatible with DeviceNet equipment from other manufacturers.
DeviceNet Cables
DeviceNet cables consist of a shielded and twisted data pair, as well as a shielded and twisted power pair for the 24 VDC bus power, with an additional outer shield. The drain wire, together with multiple layers of foil and/or braid that surround data and power pairs, create network shield sufficient to withstand harsh industrial environment. A key benefit of carrying supply voltage in the network cable is that many DeviceNet stations do
not need a further supply, allowing the user to only need to run one cable to the station. Some stations, particularly those with high current outputs, can draw too much power from the DeviceNet power supply. These stations typically have an auxiliary power connection, allowing the user to use a second power supply for just the I/O. The bus power supply still powers the DeviceNet communication electronics.
TURCK offers a complete line of molded DeviceNet™ cordsets to facilitate network installation, resulting in a faster start-up and fewer wiring errors. The bus and drop cables are specially designed foil-shielded, high-flex cables with very low inductance and capacitance to minimize propagation delay time. In most cases, bus cable connections are made using 5-pin minifast (7/8-16 UN) or eurofast (M12) connectors. A variety of stations are also available that support terminal-block type connections. TURCK cordsets for the DeviceNet system are available in standard lengths.
Network Power
The DeviceNet power may be provided by single or multiple power supply units. The nominal voltage and current rating of all power supply units are: 24VDC +/- 1%, up to 16A continuous feed. The output has to be isolated from AC and a chassis ground. It should be protected against over voltage or over current. The power supply units have a rise time less than 250ms to reach 5% of its rated output voltage. When multiple power supply units are used in an application, V+ is broken between them and only one is grounded.
Networking Grounding
The DeviceNet is grounded at a single point according to the Figure 4. The shield and the common (V-) of the power supply unit are brought to the same earth-ground point using a copper braid or #8 grounding wire. The shield must be continuous and serves as only protection against electrostatic discharge (ESD) and fast electromagnetic interference transients (EMI), the common source of network communication problems. The location of the grounding point, affects, the same way as location of the power supply unit, a quality of the CAN signal and data transmission. The middle of the network load is considered to be the best possible location for the network grounding.
A common mode voltage (CMV) also depends on the location of the network power supply. The higher common mode, more communication errors is generated. The maximum CMV is 9.3V.
Electronic Data Sheets (EDS) Files
Electronic Data Sheet (EDS) is the DeviceNet configuration file that contains information about a device: identity, I/O data size and the device’s configurable parameters. The information provided by EDS files is imported into network configuration tools and guide a user through the steps necessary to configure a device. EDS files are available on TURCK website.
The DeviceNet stations support different diagnostics information which depends on their complexity and functionality. The common ground for all devices is that they have to provide visual diagnostic information using either a pair of status LEDs called MOD and NET LEDs or use a single combined MOD/NET led. The behavior of these LED is described hereafter. TURCK discrete IO modules provide diagnostic data as a part of input data map. The standard stations support group diagnostics, where a single alarm bit is set if any IO is faulted. The deluxe station support individual IO diagnostic data, like open wire and short wire alarm bits. All TURCK devices support MOD and NET LED diagnostics as follows:
MOD – Module Status LED
Off Not Powered There is no power applied to the device.
Flashing Green Device in Standby state. Autobaud detection not completed. • Device needs commissioning due to configuration missing, incomplete or incorrect.
• Device is not in a scan list. Configure device.
• Autobaud detection not completed. Check CAN lines.
Green Device Operational Device operating normally.
Flashing Red Minor Recoverable Fault Recoverable fault. For devices with group diagnostic it indicates I/O fault. Check I/O for short.
Red Major Unrecoverable Fault The device has an unrecoverable fault; may need replacing.
Flashing Red-Green Device is powered and in self-test mode. Self-test mode during power-up sequence.

NET - Network Status LED
Off Not Powered, Not On-Line • No network power.
• Device may not be powered.
• Device has not completed DupMacID test yet.
Flashing Green On-Line, Not Connected Device has passed DupMacID test, it is online, device is not allocated to a master. Device is not in a scan list.
Green Device Operational, On-Line and Connected Device is configured, connected and communicating.
Flashing Red Connection Time-Out One or more I/O connections are in timed out state.
Red Critical Fault or Critical Link Failure Failed communication device. The device has detected an error that has rendered it incapable of communicating on the network. Bus-off.
Flashing Red-Green Device is powered and in self-test mode. Self-test mode during power-up sequence.
The DeviceNet supports 64 nodes, which are assigned addresses from 0 to 63. The address 0 is usually assigned to a scanner, addresses 1 through 61 may be assigned to different nodes, address 62 is reserved for a configuration tool, address 63 must be always free and it is reserved for node commissioning. The station’s default node address (out of box) is 63. Each node’s address must be initially set, usually via rotary dials or switches on the node. The address can also be set with a DeviceNet configuration tool. Changes to the address settings take effect when the station power is cycled. Every device supports duplicate address detection mechanism (DupMACID) that prevents multiple nodes to occupy the same node address. The DupMACID is run at the device power up. The Device shuts down (goes into busoff state) when it detects that there is another device at the same node address.
Data Rate
The data rate or baud rate is the speed of data transmission over the DeviceNet. The DeviceNet operates at three data rates: 125kbps, 250kbps and 500kbps (kilobits per second). All devices must be set to the same data rate in order to have functional network. A failure to do so may result in a critical network fault, forcing entire network or some devices to go into bus-off state. The bus-off state is a state of a device (i.e. CAN chip) when it detects an error that has rendered it incapable of communication on the network. The data rate is selected using a data rate switch. In general, the switch may have 4 positions marked: 125, 250, 500 and PGM (programmable mode).  Instead of the data rate switch, many devices support Autobaud detection mechanism that automatically detects network data rate during power up and sets the device baud rate accordingly. The Autobaud and data rate are supported by the device parameters that can be enabled/disabled and set using the device EDS file (Electronic Data Sheet).
Network Length
The DeviceNet bus uses trunk and drop topology. The trunk is the main communication cable, and requires a 121 ohm resistor at both ends. The maximum length of the trunk depends on the communication rate and the cable type. Drops are branches off the trunk, and may be from zero to 6 m (20 ft) in length. The cumulative drop lengths are dependent on the communication rate. The following table shows the maximum ratings for a trunk using the most common cable types as defined by the DeviceNet specification.
125 kbps 500 m (1640 ft) 300 m (984 ft) 100 m (328 ft) 420 m (1378 ft) 6 m (20 ft) 156 m (512 ft)
250 kbps 250 m (820 ft) 250 m (820 ft) 100 m (328 ft) 200 m (656 ft) 78 m (256 ft)
500 kbps 100 m (328 ft) 100 m (328 ft) 100 m (328 ft) 75 m (246 ft) 39 m (128 ft)