Topic: Sunrise/Sunset in 2550 Console

[tweatherman]

I've noticed that these times are off by a minute at sunrise and 2 minutes at sunset on the console compared to the web app which is correct. I double checked that the coordinates are correct in the console but still getting discrepancies with this. Anyone have a idea for a fix on this? The screenshot of my console wouldn't post do to size.

Thanks,
tweatherman

[Rover1822]

May I ask, how you know the web app is correct?

[tweatherman]

It corresponds with my other weather apps and via online.

tweatherman

[Rocketman]

What version FW are you running on the console ?

[tweatherman]

WSView app says 1.6.6 and the web app still says 1.6.5.

tweatherman

[Rocketman]

sorry, I though you were using a HP2550 as a console

[tweatherman]

I am using it as console I'm not where to view the firmware version on the console..

tweatherman

[Rocketman]

https://www.youtube.com/watch?v=gaXco_XJPgk

try this

[tweatherman]

The console showing 1.6.6 wifi firmware and the hardware firmware on console is showing 1.8.0.

Thanks,
tweatherman

[Rocketman]

latest firmware is 1.9.3 ( might want to upgrade )

[tweatherman]

I'll check it out.

Thanks,
tweatherman

[Platokidd]

Yeah, my Ecowitt consoles are always off by a minute or two depending on location (I have 3 PWS at different locations). However, my AW consoles read spot on to Weather Underground.

[tweatherman]

So this a bug on their end?

Thanks,
tweatherman

[aa6vh]

A couple of possibilities: Your console might be taking in account your altitude, whereas the web site won't necessarily know what that is. And there may be a mismatch in the definition of Sunset or Sunrise. Is it from the top of the sun, the bottom of the sun, or perhaps the middle of the sun? There can be a small (i.e. minute difference) between those datums. And what about atmosphere refraction (the bending of the light waves can make the sun visible after (or before) it sets or rises? The point is to make sure your sources are using the same algorithms.

[Gyvate]

So this a bug on their end?

Thanks,
tweatherman

I doubt that this a bug in the firmware as the firmware is the same for Ecowitt and Ambient for most of their functionality. Only for Ambient consoles a restriction to read only Ambient sensors applies. Therefore an Ambient console (WS-2000/WS-5000) should not show a different sunrise/sunset from a HP2550 (clone) console if both have exactly the same longitude/latitude data entered.

[Gyvate]

A couple of possibilities: Your console might be taking in account your altitude, whereas the web site won't necessarily know what that is. And there may be a mismatch in the definition of Sunset or Sunrise. Is it from the top of the sun, the bottom of the sun, or perhaps the middle of the sun? There can be a small (i.e. minute difference) between those datums. And what about atmosphere refraction (the bending of the light waves can make the sun visible after (or before) it sets or rises? The point is to make sure your sources are using the same algorithms.

Ecowitt (clone including Ambient) consoles do not use these specialties or different definitions, nor do they use altitude. They only use longitude/latitude for calculating the sunset/sunrise times.

[aa6vh]

Ecowitt (clone including Ambient) consoles do not use these specialties or different definitions, nor do they use altitude. They only use longitude/latitude for calculating the sunset/sunrise times.

But when comparing times to a web site, does the web site take altitude into consideration? Is the web site taking your reported position exactly, or using a generic position for the area you are in? And what is the Ecowitt/Ambient (or web site) definition of Sunset? Top, Bottom. or Middle?

I am only saying the inconsistency in the way Sunset time is calculated can give that kind of time difference.

[Rover1822]

Ecowitt (clone including Ambient) consoles do not use these specialties or different definitions, nor do they use altitude. They only use longitude/latitude for calculating the sunset/sunrise times.

But when comparing times to a web site, does the web site take altitude into consideration? Is the web site taking your reported position exactly, or using a generic position for the area you are in? And what is the Ecowitt/Ambient (or web site) definition of Sunset? Top, Bottom. or Middle?

I am only saying the inconsistency in the way Sunset time is calculated can give that kind of time difference.

Depends on the web site, as far as knowing what Ecowitt/Ambient use, no idea. But if you are looking at transmitted data from your station to a web site, the data you transmit, does not contain the sunrise / sunset values.  These are calculated by what ever math they want on the receiving end, same as your console representation using its own calculation. There are variations of the calculation for this, just as there are for other calculated values, such as dew point, feels like , heat index .. etc.  For the most part these are not considered "bugs" , just variations of calculation.

[Gyvate]

Ecowitt (clone including Ambient) consoles do not use these specialties or different definitions, nor do they use altitude. They only use longitude/latitude for calculating the sunset/sunrise times.

But when comparing times to a web site, does the web site take altitude into consideration? Is the web site taking your reported position exactly, or using a generic position for the area you are in? And what is the Ecowitt/Ambient (or web site) definition of Sunset? Top, Bottom. or Middle?

I am only saying the inconsistency in the way Sunset time is calculated can give that kind of time difference.

the console doesn't report a location to the websites (WU, ecowitt.net). Ecowitt.net takes the data that the users have provided in their accounts - if they differ from what was entered into the console, differences may occur. The console and ecowitt.net do not know each other with reference to the location entered into both.

[aa6vh]

I have a computer algorithm that I found on the net (it was written in VB, and I ported it to Javascript) that I run on my phone. It tells me, given my lat and long, where the sun is at any given time (bearing and altitude angle). I would expect similar code probably resides within the consoles. And the algorithm assumes Sunset to be when the middle of the sun reaches the horizon.

I think most people just take these trivial details for granted anyway. I suspect if you polled the population, the answer as to when Sunset is would be the top of the sun. But its only when one demands absolute accuracy that it matters.

Did you know that the earliest Sunset in Los Angeles CA happens on Dec 4, 2023, and the latest Sunrise happens on Jan 7, 2024? When I tell people this, I usually get a response of "What a minute, what about that solstice that happens on Dec 21?" (If you do not believe me, check out a table of Sunset and Sunrise times for December and January).

[tweatherman]

Shouldn't the formulas be the same in the Ecowitt app and the console? Makes sense anyway.

tweatherman

[Gyvate]

Shouldn't the formulas be the same in the Ecowitt app and the console? Makes sense anyway.

tweatherman

the Ecowitt app doesn't make any calculation - it's the ecowitt.net website which does (and the console - independently).
And - the formula used should be the same - and when the base data are exactly the same, the results should be the same.

[Platokidd]

My locations are the same for the following. My Ecowitt consoles and AW consoles have always differed by 1 minute. This holds true for aw. net and ecowitt . net.
Not going to lose any sleep over the minute, just stating a fact on my end.

 [ You are not allowed to view attachments ]

[Gyvate]

sorry to appear nit-picking, but your pictures show differences in what the websites produce and not in what your consoles produce.
To see what your onsoles produce we'd need to see pictures of your consoles, wouldn't we ? 
And - I wouldn't care about one minute in this context either ...

[Platokidd]

 [ You are not allowed to view attachments ]


ws2000 on left / hp2560 on right