DIY IR Cloud Sensor with PC Interface, any ideas and plans

[dyfed0_0]

Hi there all,

I have had a quick look on the forum and this subject is thin on the ground so I thought I would throw myself into the affray.

I have been scouring the net for a cost effective IR cloud sensor to monitor the pesky clouds, everyone's nemesis in the UK.

The cheapest commercial unit is the Auarora which is over £240 quid, then Technical Innovations which is a case of importing and then there is the Boltwood.... which is far to expensive.

Has anyone got any ideas of doing a cost effective cloud sensor with easy to follow diagrams and pictures (never really tried something like this so the electronics part is going to be my biggest nightmare....lol)

Then, to chuck more petrol on the fire.... a PC interface as well, with possibly an alarm???? to warn someone of increasing heat which would indicate the clouds were rolling in.

Any input and thoughts would be great...

Many thansk inc advance and clear skies

Dyf

[pete_l]

I've built a prototype cloud detector, which is quite successful. You can read about it here. The principle is to use an I.R. detector to remotely measure the temperature of the sky. If the temperature is low, say 25°C or more below ambient ground-level temperatures, you can say that the sky that the sensor is pointing at is clear. If the temperature the detector records is closer to ambient, then it's cloudy. I haven't got around to interfacing it to a PC or an alarm yet but other people here have done similar projects.

An example of this type of sensor (not using my design, I hasten to add) is available at the Calar Alto weather page,

[dyfed0_0]

Hi Pete,

Many thanks for that.... certainly a right step in the right direction and also looks very promising.

I have to say it is amazing how you have been able to get from a gun IR thermometer to an actual working basis that could be developed to a permanent installation.... I honestly wish I could do things like that...... unless I have an instructions in front of me, i am useless when it comes to such matters.

The PC Interface at Calar Alto looks very conscise as well..... if that the sort of PC interface are you thinking of developing.

Thats exactly what I had in mind cant wait to see your final design... all be it a non commercial enterprise.... but you could always sell your plans/ design to interested parties?? me for one

Many thanks once again Pete, if you do get a chance, would love to hear any updates done?

Dyf

[pete_l]

Dyf,

Thanks for the kind words. I wouldn't hold your breath waiting for me to get this past the prototype stage I've had to put this to one side for a while.

if you have a look on this site for cloud detectors and search through some of the Arduino threads you'll find more information from others who are trying the same things.

[yesyes]

There was a rather lengthy discussion about a cloud sensor (and motor focuser in one) a while ago here:

http://stargazerslounge.com/diy-astronomer/106144-arduino-focus-control-cloud-sensor.html

The one I build still runs as a weather station at the moment and is still recording graphs.

I haven't really put any actual cloud detection in the code but it reads sky temperature and ambient temperature and I can then "manually" decide whether the sky is clear or not. That works quite well actually.

[dyfed0_0]

I noticed the discussion about the Arduino but what put me off was the stepper control for the focuser and I would not know how to go about omitting it from the instructions.... may be a simple procedure but knowing me, I would fry something.....lol

It does look promising though as the first thread with the screen shows the cloud conditions on the right and then on the left you have focuser control... so I am sure the software is more focuser orientated..... if I am worng, please, correct me, it is one device I would love to get.

By the way, havent been here for a few months with one thing and another.... how did the experiments go for the 14" flat screen?

[ncjunk]

Yep it is both at the moment as I put the cloud sensing on hold whilst doing the focuser.

The cloud sensing software is easy to separate from the focuser side so depending on your level of software knowledge you could rewrite some of it.

If, for now, you just want the graphs and the temperature difference i could sort a version out for you (i really need to start on this again) but there is no alarming.

[dyfed0_0]

Hi Neil

I do confess, I went over to the Yahoo group and placed a similar thread up there as it is more specific question.

To be honest, when it comes to PC's, I didnt undertand BASIC let alone the newer languages.......lol I would know where to start.

A National/ European Database could then be setup, something that YESYES had suggested in the Arduino thread, monitor trends.... early warning, especially if you have an Anerometer connected up.... wind pushing clouds northerly from say France... before hitting the UK, or any where in Europe, in theory, an early warning system as well...... im going ahead of myself now......lol

If you could seperate the software to give graphs and even the block to the left of the focusing page, that would be great....... I noticed something as well, a mention of a schmatic for the cloud sensor..... havent come accross that yet... still slowly readin through the thread. If there is one, that would be great as well........

Is there a step guide as well on how to asseble it as well....lol, sorry, only joking, just pointing out how hopless I am at these things...lol but dont be suprised if I do ask for pointers though.

Many thanks once again Neil.

Dyf

[Euan]

I've got the Aurora, and what yesyes said about "manually" deciding about clear spells is about right, even for a commercial unit. Just something to bear in mind when you are making one.

Setting a level for "Clear" doesn't really work as it depends what the actual clear sky temperature is, I've noticed this change from daytime to night time. The most important thing I've found, is how constant the temperature level is.

For instance, here is an imaging run from the other night. The red level is Sky Temperature, if you monitor this and it is flat, it's totally clear outside. As soon as this gets any significant bumps, you are seeing cloud. You can see the Sky Temp gradually dropping as the night progresses, then a bump. This was thin cloud that had rolled in, it was enough to wash out the subs at that point

[ncjunk]

I agree with you Euan but you can get a base level if its clear at the start of the night and, as you have above, you will then see a significant change if a bit of cloud rolls in.

There is a bit of drift in the reading during the night obviously.

The only other problem is the possibility of dew forming on the sensor and it then reading the temperature of that.

[ncjunk]

The schematic is there somewhere...i'll dig it out later. It is one ir sensor, two pull up resistors and a capacitor.

its a simple circuit and i can help you fault find. I needed to get back to this anyway so i'll get a stand alone version done and tested....the code is all there somewhere.

[Gina]

Very interesting I'm watching this thread avidly - a nice little project for the future.

[dyfed0_0]

Hi Euan

I do have to ask this, I did a search a while back for a commercial unit and beleive it or not, the Aurora is the cheapest you can get anywhere.

Is seems to work on the same principal software and the Technical Innovations unit (all be it licenced from a company in Italy) and the of course, you have the Boltwood Cloud Sensor II where you need a second mortgage to get it.

This time round I fancied making one from scratch from the help of other users who have actually gone down the road of making one.

Anyway, going back to the start, how good is the Aurora? judging by the write up and description of it, it sounds awesome..... and if my plan of making one goes south, my next step was to invest in it..... but being a 1st hand user yourself Euan, you can make an unbiast judgement of its functions... another good thing about it is, I emailed the manufacturer and there are plans to, not link it but to have it work along side the Virtual Weather Station Software similar to the other two, more expensive options.

But I would like to have a crack at doing one first though but I agree, there will be a lot of stumbling blocks but if aired on here, the issues will get ironed out....

[pete_l]

A National/ European Database could then be setup, something that YESYES had suggested in the Arduino thread, monitor trends.... early warning, especially if you have an Anerometer connected up.... wind pushing clouds northerly from say France... before hitting the UK, or any where in Europe, in theory, an early warning system as well...... im going ahead of myself now......lol

It's a little more complicated than that, I'm afraid.

The wind at ground level can be very different (just as hot-air balloonists) from the wind at altitude: both in speed and direction. Have a look for something called the Crossed Winds rule.

Better to keep an eye on weather satellite pictures.

[Euan]

The only other problem is the possibility of dew forming on the sensor and it then reading the temperature of that.

That's true, the Aurora is heated to dry up drops on the rain sensor, it reads the ambient temperature at around +5 degrees higher than it actually is which can be a bit annoying but worth it for this reason I suppose.

how good is the Aurora?

It's pretty exceptionally to be honest. It's one of those things you buy then wonder how you managed to get by without it before. There are still a few bugs in the software, but they are sending me a beta version of the next significant release soon (actually it's over a month late...), once I get that I will post up a full review.

All I can say so far is that you wouldn't be wasting your money put it that way

[dyfed0_0]

Possibly, you could use a peltier element to heat the IR sensor but then you run the danger of getting false readings.... would have to be regulated to prevent an occurance but then, you run to the area of calibrating such a device... what fun.

Though, I am still convinced, such a device could be done if collaborated on.

I do confess, and take my hat off at the people at Aurora Tech, they have come up with a device that beats the opposition at a trully bargain price. Ill be looking forward to hear on the improvements of the new software..... as if the "VWS talk together" part of the software is included, it will be hard for me to ignore.........ahhhhh, I want to do an electronic project but the Aurora looks so good.

Why does life throw these things up.....lol

[Euan]

as if the "VWS talk together" part of the software is included, it will be hard for me to ignore.........ahhhhh, I want to do an electronic project but the Aurora looks so good

Just wait till you see the new version imitating the Boltwood text file and working with all Boltwood compatible automation programs, like ACP and CCD Commander, now you are talking

[dyfed0_0]

Imitating Boltwood now your talking because the Boltwood, from what I can gather from, it can be linked into the Virtual Weather System Software which is why I emailed the lads at Aurora to ask if they had something in the pipeline to accomodate this.....mmmmm looks like someone listened once in the manufacturing world.....lol, mind I would imagine that a few people asked for that function........ oh bother......

Im still going to do make a CS for my own satisfaction I think but for the sake of my network, the Aurora is deffinatly back on the cards (mind, I dont think it was ever off)...... but im deffinatly going to build one as well, granted, ill have two but think of the comparison experiments that could be run?

[Gina]

The price of the Aurora (£255) means it goes on the wish list well below my next scope but I can see that a cloud and particularly rain sensor would be a big benefit when I get automated image recording going. Of course, if anyone gets a home made one going, that would be a different matter - I like a good DIY build

Good luck dyfed

[dyfed0_0]

Hi Neil,

If you have a quick copy of the schematic and if possible a parts list, that would be great.

Ill leave the code in your capable hands, wouldnt know where to start to be honest, but I am sure a stand alone device could be made form your original idea.....

What are your thoughts on it? and with the issue of using a peltier element (regulated until the right setting is found) to try and prevent dew and rain giving off false readings?

Dyf

[Gina]

Actually, I'd love to add sky temperature to my weather station web site graphs.

[dyfed0_0]

Hi Gina,

I agree, the outlay for an Aurora is high, but not as high as the Boltwood for example, which is around the $1000 mark and to be honest, a complete non starter, especially when Euan has confirmed that a Boltwood emulator (good name for it.....lol) is being included in the next software release of the Aurora, well, no brainer really but I agree with you, I would love to have a go at building a CS and getting it up and running..

Granted, the Aurora is deffinatly on my shopping list, I just liek to get my hands dirty.... so to speak.....lol

The only reason why the Aurora is on my shipping list is because the VWS software which is linked to a weather station (rain sensor) , im emphasising this, they are not linked in the sense they speak to each other but it does influence the VWS readings to be more accurate, if that makes sense..... I cant remember exactly what the gentlman at Aurora said but it is something on those lines...... especially if you have an automated system.... something im not planning but if I know even a hint of a cloud over head, I can have a look and decide to close up.

The reason why I am interested in this issue..... the other night (like so many) I was sate in front og the computer doing some PEMPRO readings where the star was getting dimmer and dimmer..... looked out and it had closed up with clouds..... me being sat in my little room, didnt see it come.

If I had warning, I woud have been ready for it........ and im 100% sure Im not alone with this happening to others.

I hope the above made sense?

Dyf

[ncjunk]

I personally am not so bothered about rain. If the clouds roll in i will detect it and if it rains then i will have to close my roof and decide when to open it myself, not relying on the sensor. Dew can be prevented with a small heating element same as with the telescopes. I have a rain sensor someone very kindly sent me which has a heater on it for that purpose so i will stick the ir sensor next to that.

The ir sensor i have will be fine to give you a guide to cloud cover...you'll see the temp rise as your star slowly disappears into the gloom. What i dont have is any automation programmed yet i.e no audible alarm or setting of outputs on the Arduino.

[Gina]

Hi Gina,

I agree, the outlay for an Aurora is high, but not as high as the Boltwood for example, which is around the $1000 mark and to be honest, a complete non starter, especially when Euan has confirmed that a Boltwood emulator (good name for it.....lol) is being included in the next software release of the Aurora, well, no brainer really but I agree with you, I would love to have a go at building a CS and getting it up and running..

Granted, the Aurora is deffinatly on my shopping list, I just liek to get my hands dirty.... so to speak.....lol

The only reason why the Aurora is on my shipping list is because the VWS software which is linked to a weather station (rain sensor) , im emphasising this, they are not linked in the sense they speak to each other but it does influence the VWS readings to be more accurate, if that makes sense..... I cant remember exactly what the gentlman at Aurora said but it is something on those lines...... especially if you have an automated system.... something im not planning but if I know even a hint of a cloud over head, I can have a look and decide to close up.

The reason why I am interested in this issue..... the other night (like so many) I was sate in front og the computer doing some PEMPRO readings where the star was getting dimmer and dimmer..... looked out and it had closed up with clouds..... me being sat in my little room, didnt see it come.

If I had warning, I woud have been ready for it........ and im 100% sure Im not alone with this happening to others.

I hope the above made sense?

Dyf

Yes, I think I get the meaning As I said, I like DIY projects too. I agree about getting involved with something and not watching the weather - happened to me often! These things have a habit of taking you unawares. I would hate it if I missed a sudden shower and found rain water in my nice expensive scope. Such things are like insurance - tragedy might not happen but it's best to be prepared.

I'm thinking I might later automate my ROR but I haven't even built it yet The future looks decidedly exciting I did make a motorised garage door many years ago, operated by a magnet and reed switch as I entered the drive.

[ncjunk]

I plan on motorising my ror as well with a garage door opener. I have to sort out the track on the roof first as it sticks a bit in places which would be too difficult for the motor to move.

I would love it if i could get the scope to park and the roof to close but i will probably just put a physical button in to begin with so that i am there to supervise but dont have to struggle!