|
I had to abandon a particular approach for the sensorless, and now have a new way that is more direct, and eliminates the need for information about the rotor inductance and mutual inductance. The other way would work as long as the rotor flux angle was being set with the sensored code, but wouldn't work for very long when it would switch over to sensorless. I hope this approach works. It is much simpler, and I think it will work, and will be easy to tweak if necessary. I probably wont be able to test it today since my son has a fever and I'm getting him a regular supply of pedialytes. Once it's working I'll send over a hex file. One neat thing about it is the super fast __arctan() that I made. TI said this approach was not good since the arctan takes so long to compute. haha
|