Greetings all. I'm a power electronics engineer whose familiarity with FPGAs begins and ends with the definition of the acronymn.
That said, I would like to compare two different schemes for controlling the flux and torque in ac machines but use the exact same PWM method, namely Space Vector Modulation in both cases. SVM is rather demanding of CPU cycles if implented in software and could affect the performance of the machine control scheme, so I'd like to offload that duty to semi-dedicated hardware, hence putting it in an FPGA.
My cursory research on the subject indicates that it is certainly possible to implement SVM on an FPGA - one chap used a Xilinx Spartan 3e, hence my registering and posting here - but as to how he did it and whether it is "economically feasible" is a total mystery to me. Well, not totally true - I guess he could have used MATLAB and the VHDL option to do it, but given I haven't even so much as looked askance at MATLAB since, oh, 1993 or so, I don't think that option is a good one for me.
The exact questions I have about this sort of depend on the kind of response I get from this post. E.g. - if someone here just happens to have done this and would be willing to show/sell/license their work for a reasonable fee then I am amenable to not reinventing the wheel, etc.
Nevermind, Xilinx... Altera actually has literature on their site concerning implementing SVM (et al. motor control functions) with their FPGAs so our brief and not-so-torrid affair has come to an end, I'm afraid.