Comparison of CANopen
Software Implementation Methods
| Back to |
| CANopenIA |
| Comparison Overview |
Available CANopen Software Development Methods
a.) Do-it-yourself
This means developing the software for the CANopen protocol stack from scratch by simply looking at the specification and writing the appropriate code for the microcontroller of your choice.
Although this method allows for writing code optimized towards the microcontroller used, it is also the one with the greatest risks and an unpredictable development time.
b.) Use commercially available CANopen source code
There are several companies in the business of selling CANopen source code that supports a number of microcontrollers and CAN controllers. As the source code is already written with portability in mind, adapting the software to completely different hardware implementations is easily possible. Configuring the software to implement the required communication for a specific CANopen node and merging it with application specific code can be done within a few months.
Experienced users and consultants can sometimes get that time down to a week.
c.) CANopenIA with customized code
The CANopenIA chips provide a "COCO" (CANopen Core) library that is accessible from application specific code. Using function calls to the library, the application code can implement a completely customized CANopen node.
It should be noted that COCO indeed only offers core functionality. It implements a user-adaptable CANopen Object Dictionary and all accesses to it using SDO (Service Data Object) transfers. The processing of the more complex PDOs (Process Data Objects) is responsibility of the application code.
Depending on the experience of the user and the complexity of the CANopen node required, the development time can be in the range from a few weeks to several months.
d.) CANopenIA or CANopen daughter boards
For typical CANopen I/O nodes, the CANopenIA chips implement the entire application code on-chip as many of the plug-in/stick-in CANopen daughter boards available do.
This allows developing and building a CANopen compliant node WITHOUT writing a single line of code.
Obviously such an implementation has no software development time.