rsdoran said:
That is an improper analogy. Intel processors are used in many devices, FPGA being one of them. A regular here builds motion controllers that use FPGA.
Building a PLC is not a cheap endeavor so why reinvent the wheel? Because you can.
I'm not saying that it's impossible. It's just above the level of complexity and work required of just about anyone that isn't running a lab at Rockwell or Siemens.
Designing for an FPGA is like designing for actual silicon. You generally write up your logic in a high level HDL (VHDL or Verilog).
So, first you're going to have to figure out how you want your PLC to work internally (data files or tag based (a-la ControlLogix) memory, your entire instruction set, and how to implement complex instructions (PID for one).
Then you have to determine how to parse a user mode program (are you going to design a RLL interpreter?)
Next, you have to account for I/O
Communications are also a necessary component. Your PLC doesn't necessarily have to be network enabled, but a user has to be able to at least connect to it to download a program.
Then there are about a hundred other things that I've glossed over that make a PLC a PLC and a useful thing to have (ie. online editing)
In summary, it's one thing to implement a custom motion controller using an FPGA and quite another to try to roll your own entire PLC. A custom motion controller only requires a few specific commands as input, and the same for output (although the intervining logic can be about as complex as need be)