With SW controls, the first rule is never to command a control to move in a way ....
ErNa
Posts: 1,753
I found this in my mailbox and find it adequate to forward the link: http://www.eetimes.com/design/embedded/4206216/The-basics-of-doing-PID-design-Part-1-Classical-control-theory?pageNumber=0.
I liked this : With software controls, the first rule is never to command a control to move in a way it clearly cannot track., because that is exactly, what we often don't care about. We overstress a system and then we try to find a way around, creating additional load.
The propeller allows to do experiments easily, others can't do. Start the P, I and D part in different cogs, have them running in parallel, and than watch, what happens.
And remember: "This subject is deep and broad, and engineers spend their entire careers mastering it. "
I liked this : With software controls, the first rule is never to command a control to move in a way it clearly cannot track., because that is exactly, what we often don't care about. We overstress a system and then we try to find a way around, creating additional load.
The propeller allows to do experiments easily, others can't do. Start the P, I and D part in different cogs, have them running in parallel, and than watch, what happens.
And remember: "This subject is deep and broad, and engineers spend their entire careers mastering it. "
Comments
In all the years of reading about control systems, PIDs and all, since the 1980s and having had
a very good friend who was a graduate of control systems engineering I have never heard the
"rabbit" and "hound" thing before. I like it.
Brilliant, never heard it put like that before either.
How on earth did I get a degree in Physics (1979) and be knee deep in calculus for some years before starting that without ever coming across that use of the word?