> There is nothing as easy as the "B" device. Options available > are 1. define as a poly, 2. use a table, 3. code a bm_ plugin > in C++ (This really is easy. Tell me what you want and I will > show you.) 4. A more elaborate behavioral model as a .model > file, which becomes a plug-in.
The model is to simulate EMF induced in a generator given arbitrary mechanical input. I'm defining the mechanical position input to be a voltage across two terminals (say 1m / V), which I will differentiate to get velocity (say 1m/s / V). (Or I could feed it both velocity and position, rather than worrying about how to differentiate within my model). There are some parameters I'll want to pass defining the generator geometry (winding pitch etc..), but basically I get some "K" dependant on position, which when multiplied by input velocity gives induced EMF. (K comes from the flux integrated over a generator coil, and is basically a sinusoidal function). Future refinements of the model would include computing mechanical force as a function of current drawn from the output terminals, but right now being able to simulate the EMF would be very useful. I'm planning to build a generator model with this idealised EMF, followed by an equivalent circuit to consider the "parasitics" such as resistance, phase inductance and mutual coupling between each phase. I started digging this afternoon, and thought the VCVS model might be a useful point to start from. I then got side-tracked wondering if I should be using the .model compiler, and whether I had to make up a letter to prefix my new device with. I appreciate your help. If you want to point me at any documentation to RTFM on building models, please do so.. Regards, Peter C. _______________________________________________ geda-user mailing list [email protected] http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
