Your points 1 and 2 are correct. When you have a position address set up the times are still used to calculate intermediate positions.
Best practice is to configure everything you know about your cover. It may vary slightly on different manufacturers actuator implementations but generally it will try to do the right thing given any valid configuration.
Iāve read the thread and it might be already solved for others. Its more like two short question to be sure (I have MDT as well):
ā travelling_time_up/down will be ignored or not used if actuator provides the values?
I have configured them in HA as test case with 5 seconds each, but HA still uses the 23 secs configured in MDT actuator. Playing around with different values for up/down doesnāt changed the behavior, only MDT actuator values in ETS had an effect. So Iāve removed them from yaml, might this result in problems?
ā does HA try to write the calculated āposition valueā to the position_address?
If yes, who will win? HA or MDT actuator whoās also using it? Or will it be ignored from HA, cause actuator is using it?
Imo I should just use the integration without the traveling times, cause it looks like they arenāt used. (Version of actuator, ETS, HA, etc. is latest possible).
If this would be correct with the 25 seconds, I would assume, that my 55 seconds KNX time for a specific cover should stop after 25 seconds, if travelling_time_ values arenāt configured, like they are at the moment (not configured).
But the cover closes as expected over HA without these values. Donāt get me wrong, I like it, but I wanna understand it.
If I tell HA to use it 5/5 seconds for up and down it ignores it. Only KNX values will be used. Is this correct or am I doing something āwrongā?
How a tiny little word (precedence) can be overlooked more than once ā¦
Thx for the quick response and explanation. It works as expected for me without the time lines in yaml. Still great to have them in case you donāt have an actuator with this feature.