Closed McGiverGim closed 6 years ago
I've already considered this, but not properly implemented yet. Just getting the crossing point works, but is probably not the best. Furthermore I do not want people to chase a single number to the third digit, but to look at the whole picture that might be a bit more complex. In your case for example there seems to be an underlying feedback/resonance at maybe 30-40hz. Due to the periodic/continious nature of this superposed signal, the result can have an "instantaneous" component which is not real. In your case visible as this first "bump" in the blue line. >500 there is no feedback happening (or relatively small) and thus the bump is also missing.
Thanks for the input, it's definetly on my list! But I have to do some more tests and and come up with a satisfying solution first. :)
I think I have too much D, that is the culprit of this line aspect. But my not too powerful motors and heavy props need it to stop the rolls in a clean way.
Thanks for your feedback!
I don't know if it is feasible to add the time that the quad needs to arrive from 0 to 1 in the strength vs response time graphs, for both curves. More or less we can now the value, but to compare two graphs a precision number will be better. Maybe next to the graph and not inside?
Thanks for your great app!!!!