Hi,
When I use the dl_saving in the LB Day Solar Information component, it seems to shift an hour backward, instead of an hour forwards?
kr,
Welcome @Thomas to the forum,
This is the expected behaviour: one hour backwards. One hour less to sleep.
-A.
Thanks!
Maybe it’s just different in my country, daylight saving shifts an hour ahead instead of backwards.
Alas, the noon_alt does also give a different altitude, while, daylight savings or not, the altitude of that day should stay the same?
A, thanks for bringing this up, @Thomas.
You are right there is a bug here. The LB Sunpath is correct as well as all components where the HOY is an input since the assumption is that you’re plugging in datetimes in daylight time and not standard time if you’ve connected some dl_savings_.
However, for components that output datetimes like the the LB Day Solar Information component, it is treating the output time as if it were an input in daylight time - shifting the input backwards to try to coordinate it with standard time rather than shifting it forwards as it should be. I can see this pretty clearly if I recreate the sunrise that happened in my part of the world today a little bit before 6:20 AM Pacific Daylight Time:
Give me a little time and I’ll push a big fix.
Also this is not fully correct:
Noon is going to be different depending on whether you are using solar time or standard time. So the solar altitude at noon is going to change.
If what you are trying to get is the altitude of the sun at the highest point of the day, you should not connect any dl_savings_ and you should set solar_time_ to True. That way, you are looking at the altitude at solar noon which, by definition, it the point in the day where the sun is highest in the sky.
Ok, it seems it was a typo that was specific to the method used by the LB Day Solar Information component. I just pushed a fix for it:
… and I verified that the “Day Solar Information” component now produces the correct result (it’s matching my sunrise today to the minute).
The fix is now available via the LB Versioner.
Thanks again for reporting, @Thomas!
1 Like