While I don’t do a lot of unattended imaging, there are nights when I would like to start a long imaging run and then go to bed. I know SGP will park the scope when the sequence completes. However, there is always the issue of clouds rolling in or high winds developing and either condition warrants an end of sequence action be taken.
I know a lot of imagers buy expensive weather stations for handling these situations but it occurred to me that PHD2 Guiding knows if it gets cloudy (“star lost”) or if there are numerous, large guiding corrections (windy). So, could some future version of PHD2 Guiding have configuration options to treat these cloudy and / or windy conditions as if it were a safety device and report an unsafe condition to SGP?
I use PhD as a safety device. I do have a weather station which will shut things down when it detects cloud, but as sometimes happens it clears 10 mins later, so I set a high tolerance for cloud and let PhD do the work x that way I can use recovery to test every 10 mins for 1/2 hour or do, if clear, carry on, if still cloudy shut up shop.
That’s a really good idea, and doesn’t need any change to SGP or PHD2 at all. All that’s needed is to implement a safety driver that connects to and monitors PHD2. If PHD2 reports star lost the the safety state become unsafe. SGP monitors this in the usual way.
There are a few things that need to be resolved, for instance PHD2 may report star lost because of a slew, but as using PHD2 is such a sensitive indicator it may be possible to be fairly relaxed about reporting unsafe.
I have another suggestion to further improve this idea. When PHD2 loses the guide star if it could “auto select star” as a first port of call before starting to try and find the current star that would be great. I have done this manually on many occasions while watching the guiding during intermittant clouds, and it works very well. Often when the current guide star is lost there are other stars available on the screen. By auto selecting a star when it loses the curent star, if there is another one available it just jumps to that star and keeps guiding. It also works in reverse, when the thin clouds clear and the current star becomes saturated, in this case it jumps to a non saturated star and keeps guiding. Saves a lot of hassles with the mount jumping around looking for the current star.
But yes, I’ve used PHD2 as my test too. The downside is that by turning off recovery mode, you lose a night to an errant cloud. Personally, my favorite dector isn’t a cloud sensor (it’s fickle) but the rain sensor on my foster astro alert. It’s pretty amazing (look up Hydreon rain sensors).
The one downside to simply invoking recovery mode when the star is lost, is that if recovery mode lasts 30+ minutes, the amount of drift might need a new slew, center and auto focus routine before restarting auto guiding.
I don’t have a good feel for how often it clouds up only to clear again shortly thereafter. For now, I will use recovery mode options to simply end the sequence after 10 minutes of star lost. If its been gone for 10 minutes, it’s probably a deteriorating night anyway. As far as windy, I can use the option to throw away the subs with large guiding deviations – all you loose are some frames you would not have gotten if you shut down anyway.
In some future release, it might be useful to have an “OnRecovery()” function that could specify what you would like to do before restarting the subs. I can understand someone not wanting to lose an entire night when there are some brief clouds at 9PM.
My cloudwatcher and solo arrive tomorrow. I’m excited, but I’m trying to think “HOW” can temperature detect clouds? Day/night - temp has to be all over the place? I saw the sky camera cloud monitor that is amazing.
I bought this program but finally abandoned it. I could not get it to work reliably with an SX Superstar camera and a wide angle lens (basically - an Oculus). The developer was very supportive but, at the end of the day, I could not overcome the problems (mostly associated with ASCOM) and I stayed with the AAG as a more reliable monitor, coupled with a Hydreon rain detector (that is very sensitive).
It looks at deep sky IR temp vs. ambient temp. When there is cloud cover, it reflects outward IR and the ‘sky’ temperature is a closer match to the ambient temperature. When there is no cloud, the sky temperature drops with respect to the ambient. An equation works it all out. There are a number of different ways of doing the same thing (if you like a project) including measuring the output of a Peltier (as used in cooled CCD cameras), with one side exposed to the sky and the other shielded from it. The current generated by the difference between the surface temperatures is an indicator.