Hi @Cobra. Thanks for all your work on this driver. I've forked your repo and made some custom attributes for temps, wind, and such that are a bit more friendly for Dashboard tiles.
I've also done some code cleanup and added a 1 minute auto poll option. Please feel free to check it out. And if you'd like to integrate my changes back into your version, I'd be happy to submit a pull request.
I've pushed a small update (4.6.2) to my fork to support tempHandler for apps like the built in Notifications app (pilfered from the NST Manager thermostat driver). As the code was before, Notifications wasn't able to get the temperature from the driver.
I wanted to check in here. I have not paid close attention to the data for the most part, so I am not sure when this changed - or if this is even changed or not! However, the illuminance reporting seems to be off somehow.
My device, just force polled, reports 293 (which by the way is exactly the same as 'solar radiation' being reported - something I am just noticing).
Yet if I review my data at ambientweather.net right now, I am getting 33k (mostly bright and sunny here at the moment).
Is the Illuminance reporting off here?
My inside lights are set to go off @ 1000 lumens, but have been on for 2 days.
Just checked my device illuminance against my station from which the data comes from and its spot on. data is watt meters which is different measurement from lumens
Thanks. Are watt meters just the same as watts? So if I want about 1000 lumens, just convert that to watts? Which is about 67 or so watts?
Odd, with a slow starting weekend, I noticed the lights staying after it was plenty bright out - I am usually out and about or off to work on weekdays. But I've felt the automation write as it currently is, has worked without issue previously. I guess I am wrong about that.
From my knowledge they are different measurements. When I setup my lighting I took the unscientific approach. When it seemed dark enough for lights I looked at the solar radiation number and setup my automatons based on that reading. Same method was applied for morning light.
One thing to be aware of; if the station reporting the illumination data is delayed sending, and/or wunderground is slow on reporting you can have issues when the actual light intensity is not matching the reported.
For a more granular setup you could use a lux sensor such as Aeon Multisensor 6 to get your light readings from. However that measurement is in lux, not watt meters. Its not a big deal just different numbers. Using a sensor like this in your setup eliminates the extra complexity of using cloud data.
The GitHub link is dead. The referral to the CobraVMax site indicates that no login is available yet as the site is still under construction. So, this appears to be a dead end.
I'm having problems with the refresh rate. I have it set to poll every 10 minutes during the day and every 3 hours at night. It polled at 6:48 AM, and then not again until I finally pushed the poll data and refresh buttons at 6:07 PM. Is there a limit or issue with the external data or something else that would cause this?
ETA: It does seem to be updating lux every 10 minutes now, and that's what I care most about anyway. But it didn't update lux either between 6:48 AM and 5:08 PM, so something seems to be going on.
It was not in night time mode.
My location data is correct.
It actually seems to be refreshing every 3 hours at night, as I have it set to do. But during the day, it just doesn't refresh. Today there were no events between
6:53 AM and 6:34 PM.
It's during the day that I really need more regular data.
I'm not sure what to try. I suppose I can go back to the dark sky driver, but that solution has a limited shelf life.