1.1.5 CAN and DeviceNet
Only the definition of how to detect errors, create and transmit a frame, are not enough to define a meaning for
the data transmitted via the network. It is necessary to have a specification that indicates how the identifier and
the data must be assembled and how the information must be exchanged. Thus, the network elements can
interpret the transmitted data correctly. In that sense, the DeviceNet specification defines exactly how to
exchange data among the devices and how every one must interpret these data.
There are several other protocols based on CAN, as DeviceNet, CANopen, J1939, etc., which use CAN frames
for the communication. However, those protocols cannot be used together in the same network.
1.2 DEVICENET N ETWORK CHARACTERISTICS
Introduced in 1994, DeviceNet is an implementation of the Common Industrial Protocol (CIP) for industrial
communication networks. Developed originally by Allen-Bradley, it had its technology transferred to the ODVA
that, since then, keeps, publishes and promotes DeviceNet and other networks based on the CIP
protocol.
Furthermore, it uses the Controller Area Network (CAN) protocol for the data link and access to the medium,
layers 2 and 1 of the OSI/ISO model, respectively.
Used mainly for the connection of industrial controllers and I/O devices, the protocol follows the model
producer-consumer, supports multiple communication modes and has priority between messages.
It is a system that can be configured to operate in master-slave architecture as well as in a distributed point-to-
point architecture. Besides, it defines two kinds of messages, I/O (process data) and explicit (configuration and
parameter setting). It also has mechanisms to detect duplicated addresses and for node isolation in case of
critical faults.
A DeviceNet network can have up to 64 devices, addressed from 0 to 63. Any of them can be used. There is no
restriction, although the 63 should be avoided because it is usually used for commissioning.
1.3 PHYSICAL LAYER
DeviceNet uses a network topology of the trunk/derivation type that allows the signal wiring as well as the power
wiring to be present in the same cable. This power is supplied by a power supply connected directly to the
network, which feeds the CAN transceivers of the nodes, and has the following characteristics:
24 Vdc;
DC output isolated from the AC input;
Current capacity compatible with the installed equipment.
The used Baud rate depends on the size (cable length) of the network, as showed in the table 1.1.
Table 1.1: Network size x Baud rate
Baud Rate
In order to avoid reflections in the line, it is recommended the installation of termination resistors at the line
extremes, because the absence of them may cause intermittent errors. This resistor must have the following
characteristics, according to the protocol specification:
121 Ω ;
0.25 W;
1% tolerance.
CIP actually represents a family of networks. DeviceNet, EtherNet/IP and ControlNet use CIP in the application layer. The difference among
them is primordially in the data link and physical layers.
Phone: 800.894.0412 - Fax: 888.723.4773 - Web: www.clrwtr.com - Email: info@clrwtr.com
To Next Page
Loading...
