FP to cRIO with software developed in Borland C++ for the last 20 years

[Pedcast]

Dear community.

I would like to know if you can help me with the following.

I have the need to migrate a FieldPoint architecture to CompactRIO, the problem is that the control software of that FP architecture is 20 years old developed in Borland C++ running on a windows 2000 computer, with some classes that communicate to the FP1000/1001 via serial with ASCII commands, to interrogate the modules (FP-AO-200, FP-AI-100 and FP-RLY-420) and get the signals and control the system.

With the cRio9048 for example, can I still use the same control software and somehow generate a data driver in the FPGA to ANSWER the INTERROCTIONS of the Borland C++ application in such a way that the communication is transparent?

Thank you very much.

[James_McN · Accepted solution]

There are a few pieces to this. The short answer is yes but there are some decisions to be made.

Caveat: I'm not familiar with borland C++ so I'm going to assume getting that to GCC is feasible.

So you can compile C/C++ to the crio using the cross compile chain for NI LinuxRT.

However the IO model for cRIO is very different. In LabVIEW there is the scan mode which is more similar to FP but this isn't available in C as far as I can find.

You can create a very basic FPGA VI and then use the FPGA C interface to generate bindings you can call from C. 

Personally I would also question whether cRIO is the right migration platform. You could also look at some kind of modbus IO platform which has a more similar access model to FP and you then don't have the FPGA to deal with. Assuming you want a like for like replacement the FPGA just represents extra time and money as you will need LabVIEW and LabVIEW FPGA to program it. Or find an NI partner who can create the wrapper for you.

Much of this depends on the bigger picture, but I hope that gets you started!

[Pedcast]

@James_McN

I understand.

Thank you very much.