WEG DeviceNet CFW500 - Introduction to the DeviceNet Communication; CAN

1 INTRODUCTION TO THE DEVICENET COMMUNICATION
In order to operate the equipment in a DeviceNet network, it is necessary to know the manner this
communication is performed. Therefore, this section brings a general description of the DeviceNet protocol
operation, containing the functions used by the CFW500. Refer to the protocol specification for a detailed
description.
1.1 CAN
DeviceNet is a network based on CAN, i.e., it uses CAN telegrams for exchanging data in the network.
The CAN protocol is a serial communication protocol that describes the services of layer 2 of the ISO/OSI model
(data link layer)
1
. This layer defines the different types of telegrams (frames), the error detection method, the
validation and arbitration of messages.
1.1.1 Data Frame
CAN network data is transmitted by means of a data frame. This frame type is composed mainly by an 11 bit
2
identifier (arbitration field), and by a data field that may contain up to 8 data bytes.
Identifier
8 data bytes
11 bits
byte 0
byte 1
byte 2
byte 3
byte 4
byte 5
byte 6
byte 7
1.1.2 Remote Frame
Besides the data frame, there is also the remote frame (RTR frame). This type of frame does not have a data
field, but only the identifier. It works as a request, so that another network device transmits the desired data
frame. The DeviceNet communication protocol does not use this type of frame.
1.1.3 Access to the Network
Any device in a CAN network can make an attempt to transmit a frame to the network in a certain moment. If
two devices try to access the network simultaneously, the one that sends the message with the highest priority
will be able to transmit. The message priority is defined by the CAN frame identifier, the smaller the value of this
identifier, the higher the message priority. The telegram with the identifier 0 (zero) is the one with the highest
priority.
1.1.4 Error Control
The CAN specification defines several error control mechanisms, which makes the network very reliable and
with a very low undetected transmission error rate. Every network device must be able to identify the
occurrence of these errors, and to inform the other elements that an error was detected.
A CAN network device has internal counters that are incremented every time a transmission or reception error is
detected, and are decremented when a telegram is successfully transmitted or received. If a considerable
amount of errors occurs, the device can be led to the following states:

Error Active:
the internal error counters are at a low level and the device operates normally in the CAN
network. You can send and receive telegrams and act in the CAN network if it detects any error in the
transmission of telegrams.

Warning:
when the counter exceeds a defined limit, the device enters the warning state, meaning the
occurrence of a high error rate.

Error Passive:
when this value exceeds a higher limit, the device enters the error passive state, and it stops
acting in the network when detecting that another device sent a telegram with an error.

Bus Off:
finally, we have the bus off state, in which the device will not send or receive telegrams any more.
The device operates as if disconnected from the network.
1
In the CAN protocol specification, the ISO11898 standard is referenced as the definition of the layer 1 of this model (physical layer).
2
The CAN 2.0 specification defines two data frame types, standard (11 bit) and extended (29 bit). For this implementation, only the standard
frames are accepted.
Phone: 800.894.0412 - Fax: 888.723.4773 - Web: www.clrwtr.com - Email: info@clrwtr.com