search

peterhinch / micropython-samples

Assorted code ideas, unofficial MP FAQ, plus index to my other repositories.
MIT License
442 stars 91 forks source link

astro, sun_moon math #37

Closed permaBox closed 6 months ago

permaBox commented 7 months ago

There is two issues with the moon data:

  1. Set/rise time is plus one hour, probably a time zone issue
  2. Rise time has low precision, roughly plus 8 minutes

Used location: Ulm Germany, 48.4010822/9.9876076 https://aa.usno.navy.mil/api/rstt/oneday?date=2023-12-03&coords=48.4401822,9.8976076&tz=1&dst=false

Sun data look ok!

peterhinch commented 7 months ago

Thanks for the report. I am actively developing this at the moment, so please grab the latest version and re-test.

What time zone are you in? If UTC+1, you need to set the local time offset to 1. However this will affect sun and moon data. Please let me know if this fixes problem 1. I will investigate problem 2.

peterhinch commented 7 months ago

I have pushed an update with the latest improvements from Raul Kompaß. Here is the outcome on a Pico W (the platform with the poorest precision):

>>> from sun_moon import RiSet
>>> rs = RiSet(48.4010822, 9.9876076, 1)  # Ulm Germany
>>> rs.sunrise(2)
'07:55:48'
>>> rs.sunset(2)
'16:26:04'
>>> rs.moonrise(2)
'00:29:49'
>>> rs.moonset(2)
'13:27:52'
>>>

Times from timeanddate.com (only 1 minute precision): Sun 7:55 - 16:25 Moon: 00:29 - 13:28

Interestingly a 64-bit platform now produces exactly the same results.

Note that .set_day() and the RiSet constructor set the object's date relative to the system clock. The latter must therefore be initialised to local time. It doesn't need to be accurate so long as the date portion is correct when called. I will clarify this in the docs.

peterhinch commented 6 months ago

I assume this is now fixed, but please check for updates. I have improved the moon phase calculation, with an API change.