Back to overview

CANopen Networks Ensure Dependable Data from Pressure Sensors

CANopen Networks Ensure Dependable Data from Pressure Sensors

Linking Pressure Sensors and Other Equipment in CANopen Networks Enhances Production Processes

Getting networked hardware to communicate seamlessly can be a challenge if you have incompatible equipment. To address this, WIKA and other suppliers offer pressure sensors that are certified for CAN in factory automation.

CANopen is a communication protocol and device profile specification for networked equipment; the word “open” in the title refers to the open, interlinked capability of such networks. The CAN in Automation international users’ and manufacturers’ group (CiA) is the industry organization that sets the standards to ensure compatibility in production environments.

Generally, a CAN network consists of a linear bus with termination resistors at both ends. Data and power supply lines are combined in one shielded CAN cable, which is routed from one bus participant to the next. This type of connection is achieved via products designed with integrated Y-junction boxes or via external T-pieces. Not only does using CAN in factory automation eliminate the requirement for thick wiring harnesses, a CAN network’s design eliminates the need for additional cabling when the overall network must be expanded.

ABI, a leading manufacturer of construction machinery used in civil engineering, is an industrial company that puts CAN networks to good use. In state-of-the-art machinery, networked pressure sensors replace analog pressure sensors; CAN networks contain angle encoders, inclination, I/O modules, and CANopen pressure sensors used for valve control feedback. These pressure sensors measure, record, and control working pressures during piling and drilling processes. Logging such data is critical out of concern for nearby communities and buildings. In most situations, requests for bids spell out these data-logging requirements.

Because of the high shock and vibration associated with CAN network projects, as well as limited mounting space, ABI opted to connect its devices via short stubs instead of the conventional T-piece. In principle such stubs are no problem as long as they do not exceed a maximum length – if they are too long, bus traffic might be compromised. This principle underscores the need to be mindful of bit rate when configuring a bus: the longer the line, the lower the necessary bit rate.

At a bit rate of 10 kbit/s, the maximum bus length is 1 km. In CAN networks with short bus lengths, the bit rate can be up to Mbit/s. In order to ensure the best possible resistance to interference, the cable shield is coupled to the sensor’s housing through the connector. Digital data transmission avoids the line-born signal disturbances found in analog systems. All devices should be connected to an equipotential ground, if possible, and additional galvanic isolation is recommended if potential differences exist (such as distributed measurement points).

Distributed intelligence gives these pressure sensors in a CAN network additional functions besides measurements. Every CANopen device features an object dictionary, which describes its complete functionality and can be read and, to a limited extent, altered through indices and sub-indices. Section I, the CiA communication profile section, includes basic information such as device type and designation, hard and software version, error status, and CAN-identifier used. For pressure sensors, Section II includes such distributed intelligence features as information on pressure ranges, pressure units, calibration functions, filter settings, and other measurements.

Unlike the previous two, Section III, the manufacturer-specific section, is not standardized. This means manufacturers of CAN network devices can implement their own distributed-intelligence features—such as in-house service functions, intelligent/diagnostic functions or enhancements—that are not covered by the profile.

In order for the devices to communicate distributed intelligence information, all bus participants operate at the same bit rate but with different node identifiers (node-ID). In principle, there are three ways to configure the node ID and bit rate: via a DIP switch, by changing the entry in the object dictionary, or via layer setting services (LSS) services. The DIP switch configuration has the advantage that additional hard or software tools are not required to reconfigure or replace a device. ABI’s machines operate in very harsh conditions, so the company chose sensors with an all-welded construction. This construction made the DIP switch option impossible for CAN in automation at ABI.

CiA DSP 305 standardizes a procedure for setting node ID and bit rate. When LSS are used, the devices are not addressed with LSS address; instead, the device address consists of a manufacturer number, serial number, product number, and version number. With these four parameters, all CAN network devices can be identified worldwide. However, this relatively complicated procedure is best used in cases in which the configuration of the plant changes frequently, such as for engine test benches. Distributed intelligence in this case means that the LSS addresses of the devices are stored in a database. From it the user selects the devices they use; the software then assigns the addresses automatically.

The third configuration option is to change the respective entry in the object dictionary. To achieve this, the corresponding entry can be queried and changed, if required, by means of any CANopen software available on the market. A message called a service data object (SDO) is used to read or change the entries in an object dictionary. SDOs are always confirmed messages — i.e. the receipt of each message is acknowledged by the receiver. The message indicates the index and sub-index of the respective object, the access mode (read-write), and service data. The object dictionary also contains the process values, which can be accessed via SDO.

In practice, however, measured values are transmitted via process data object (PDO) messages. These distributed intelligence messages can be automated at regular intervals, in emergency messages, or for other special situations. CAN in automation offers two monitoring possibilities: node-guarding, which must be answered within a specified time; and heartbeat, which sends messages automatically at defined intervals. 

The reduced wiring requirements, high system flexibility, standardized communication mechanisms, and distributed intelligence within sensors are all great reasons for changing from conventional analog sensors to a CAN network. Besides construction machinery, other applications for CAN-compatible pressure sensors include:

  • Commercial vehicles
  • Machine-building industries
  • Medical engineering
  • Mobile cranes
  • Power-generation equipment
  • Trains and mass-transit vehicles

Not all device manufacturers have implemented all of the functions described earlier, such as node guarding and LSS. Be sure to carefully consider your requirements, then select devices that will both meet your organization’s needs, conform to corresponding CiA standards—and offer the reliability needed to make the most of CAN in automation.

CANopen Networks Ensure Dependable Data from Pressure Sensors