Sky condition system for potential automation.... what is there?

[swag72]

I am looking at a sky condition system for potentially automating my observatory. So far I have found the AAG Cloudwatcher and that is about that...... Having read some of the stuff around the AAG about their new rain sensors they are having to use, manual calibration for rain, humidity sensors....... I'm not sure that is the system that I want.

What else is out there? What are you using?

[alan4908]

32 minutes ago, swag72 said:

I am looking at a sky condition system for potentially automating my observatory. So far I have found the AAG Cloudwatcher and that is about that...... Having read some of the stuff around the AAG about their new rain sensors they are having to use, manual calibration for rain, humidity sensors....... I'm not sure that is the system that I want.

What else is out there? What are you using?

I'm using one of these http://www.iankingimaging.com/show_product.php?id=1720 to automate my observatory - a Hitec astro product which I've found it very reliable. I use it in conjunction with the Hitec astro RoR controller, so that when it is too cloudy or rain is detected, the telescope is automatically parked and the roof is closed.

Alan 

[swag72]

Doh! Of course I'd forgotten about Hitech Astro!! I don't believe that this system can be used to reopen roof's though? Also there's no wind measuring device with it I think?

[alan4908]

16 minutes ago, swag72 said:

Doh! Of course I'd forgotten about Hitech Astro!! I don't believe that this system can be used to reopen roof's though? Also there's no wind measuring device with it I think?

It can be used to automatically reopen roofs (which I do), however, it doesn't have a wind measuring device.

Alan

[swag72]

OK thanks! I'm currently investigating this from the US..... I know someone who uses it and rates it highly. I do want a wind measuring thingie!

[alan4908]

1 hour ago, swag72 said:

OK thanks! I'm currently investigating this from the US..... I know someone who uses it and rates it highly. I do want a wind measuring thingie!

Looks interesting, however, it doesn't seem to have the ability to open/close the roof based on cloud or rain.

Alan

[swag72]

It does link into SGP so will close when unsafe conditions are detected..... As far as I know, the stumbling block at the moment is SGP that doesn't then allow a roof to re-open and a sequence to restart on safe conditions...... Perhaps someone can confirm if this is the case.... 

Didn't make myself clear..... I'll be getting Skyroof too..... so should be all covered 

[andrewluck]

I use one of these: http://www.auroraeurotech.com/CloudSensor.php

Works well with CCD Commander to open and close the roof. No wind speed measurement though.

Andrew

[swag72]

It looks like if I want wind conditions it really narrows down my choices

[Oddsocks]

2 hours ago, swag72 said:

It looks like if I want wind conditions it really narrows down my choices

That's what I found too Sara.

A few observations for you below.....

If you want accurate wind speed, not only for observatory monitoring but also for general weather records then you need a mechanical anemometer, like the type offered for the AAG unit. The thermal anemometers that use no moving parts are not accurate at low wind speeds and vary greatly in reported windspeed with changes to rH. They are only useful for detecting and closing the observatory in high wind speeds.

My experience with one of the earlier AAG Cloudwatcher units has been fine. 

Initially I was using it with the AAG Solo. 

The Cloudwatcher was mounted up on one of the roof chimney stacks, sharing the TV aerial mast, and cabling it to the observatory would have been impossible with USB alone. I needed the SOLO for the network function so the Cloudwatcher was connected to the SOLO and the SOLO was connected to the house LAN.

In practice the chimney stack mount proved unreliable, although the chimney had long since seen a real fire, there was sufficient hot air wafting up the chimney in winter from the central heating and that threw the IR sensor of the Cloudwatcher right out of kilter and it would often report cloudy when it was clear.

Then there was the bird-poo problem, where the rain sensor, light sensor or IR sensor would become contaminated with droppings, stop responding, and require a climb up a ladder to clean it. I built and installed a remote high pressure spray, with two headlamp spray heads and a pump salvaged from a car scrap yard for £10, this worked in the summer but was frequently frozen in the winter. I did begin a project to add heaters but decided it was all becoming overcomplicated so I dismounted the Cloudwatcher from the chimney last summer and it is now on a short mast next to the observatory where I can reach it without ladders for cleaning, and directly connected to the observatory computer so the SOLO is unneeded and mothballed. It was difficult though to find a location near the observatory that was not liable to pick up spill light from surrounding buildings and report 'Light" when it was really dark. Dust and pollen accumulation on any of the sensors will throw the readings off so planning the location for ease of cleaning is just as important as location for best sensing.

I found that a wipe over of the sensors every few months with the silicone based car window rain repellent 'Rain-X' helped to stop pollen and dust sticking to the sensors and aided self-cleaning when it rained hard.

In use, the AAG Cloudwatcher IR cloud sensor needs a little software threshold adjustment as the seasons progress and atmospheric average temperature change. 

Situated right on the coast I see huge swings in sky temperature almost hourly as wind direction changes from on-shore to off-shore so it takes a little time to set up accurately but this last winter season I seem to have the settings nailed down and the AAG did not misreport once. I did begin a software project to read my local airport automatic weather stations every hour and automatically update the AAG settings accordingly but found out after several months work that all my local airports stop updating the automatic weather reports when the control towers close for the night and they just repeat the last report before closedown every hour until they open again in the morning!

Maybe I will return to the project and use the daytime reports to create a trend for auto adjustment of the AAG settings but really this is just a 'fun' project, it is not needed for day-to-day operation of the AAG.

I am surprised that the AAG Cloudwatcher seems to gained a reputation on some forums for being temperamental, this has not been my experience and I wonder if that is just because Lunatico Astronimia has been more open on their Yahoo group message board than other manufacturers have been? Certainly all cloud sensors that use thermopile IR sensors will need some adjustment to accommodate seasonal changes. I know of two Boltwood owners here in Dorset that have had to return their units after just a few years use to have the thermopile windows replaced and this seems to be an accepted running cost with these units (apparently they are intolerant of airborne salt so not a good idea if you live on the coast).

While replacing an IR thermopile in the AAG or Skyalert seems fairly inexpensive I'm told the last Boltwood that went back to Canada for the thermopile replacement cost nearly £800 by the time tax, courier and insurance charges were included.

The rain sensors used in the AAG Cloudwatcher and HiTech Astro are very small in surface area and very often during the daytime with the observatory closed I notice it can be raining for quite some time before a single drop lands on the relatively small rain sensor. I see the rain sensor only as a failsafe backup to the cloud sensor, as long as you accept this and ensure your observatory can withstand a light sprinkle of rain before the sensor detects water and the roof closes then all will be fine. That means making sure all AC mains power points and AC/DC convertors are suitably sheltered.

I have noticed than in thunder storm conditions, where lots of upper atmosphere cold air is being drawn down around the storm cell that the IR cloud sensor does not detect the thunder cloud until it is right overhead but wind blown rain can be falling quite heavily before that happens....

As my observatory dome controller has a direct hard-wired trigger input to close the roof I have used the AAG's built in rain sensor relay contacts to provide an independent close-shutter command to the dome, so even if the software or computer goes AWOL the shutter will still close if rain falls.

Being a bit paranoid about these things I also have an independent rain sensor wired in parrallel and using the same switch input to the dome controller to close the dome. The independent rain sensor uses the same technology as the automatic windscreen wipers in modern cars, detecting rain droplets directly by reflected infra-red off a transparent surface http://www.weatherstations.co.uk/hydreon.htm these need placing below the minimum acquisition altitude as they emit quite a bit of IR which you don't want to impinge on the optics.

As for the observatory software I began automation like Steve Richards (Steppenwolf) discounting SGP and using CCD commander as it can monitor weather and restart the series later when weather improves.

After a while though, with a dozen or so weather interrupted and partially completed targets in CCD commander, I decided to move to ACP as the observatory controlling software, with the killer function for me being ACP's scheduler that keeps track of incomplete targets and fits them in to any available gap with good target preconditions (moon phase, elevation, seeing, etc etc). Since installing ACP and even given our fickle UK weather my acquired data and completion rate for targets has more than tripled. I know this can be done in scripting with other packages but I am far too lazy these days to want to go down that path. As long as I try to forget how much ACP and ACP scheduler cost then I keep smiling 

William.

[swag72]

Thanks William for your very in-depth reply......... It looks like the AAG really is the best all rounder, epecially with the wind measurements that I really do want. I was initially looking at roof automation to run with this, but now I'm not going to bother.......

I am interested why you've not kept the Solo up and running? I would have thought this was ideal as I don't want to link the system to my main PC and so this was something that I was considering imperative to the whole thing. Location information from you is great..... it further confirms where we were going to put it (slightly above the observatory) as the correct place for so many reasons. 

[Oddsocks]

3 hours ago, swag72 said:

I am interested why you've not kept the Solo up and running?

Purely to keep the observatory systems as simple as possible, being an old engineer you learn that the more complex systems become the more likely they are to fail!

We are often away for lengthy periods and I operate the observatory remotely, though in truth ACP runs the observatory, I just collect the latest images it puts in my Dropbox once a target series is complete.

There is one full time, never switched-off, industrial computer dedicated to running the observatory, with 4xUSB3 ports, 4xUSB2 ports and 6 RS232/422 ports built-in, all the equipment can be connected directly with no hubs, (though I do use a pair of Startech USB2 powered active extenders for the SX Lodestar and QSI 683 as the cable runs for these were a little over 14ft), having only one computer to monitor and reboot remotely in case of software problems seemed a simpler approach than possibly having to remote into the SOLO as well.

The basic AAG software running on the observatory computer is enough to give ACP all the weather input it needs to control the dome so for now the SOLO is not being used, though I may go back to it in future if/when we relocate and I am away from home less often.

For the twelve months the SOLO was in use it worked without issue though I did lose weather reporting and six weeks of imaging halfway through a three month trip away from home when the broadband fibre router locked up and there was no way to remote in to the frozen router to reset it. ACP could have continued with its programed target list but once the network went down and it was no longer able to read the SOLO weather files it just closed up the observatory and waited for me to return home and fix the router.....needless to say, the old router went in the scrap bin and a more reliable router is now installed.

William.

 

[swag72]

Thanks for the info William - Most interesting and thought provoking