I'm interested to know how the rain sensor has performed for you in real-world use.
Most commercial setups have some kind of heating setup to evaporate the water. Have you found that to be an issue with yours? I found someone selling a similar sensor with a built in heater (link is on an iPad somewhere around here) which would enable the heating portion to be managed.
I need to put the box about 5-10m away from the observatory. How are you powering/running the comms? Can the entire comms/power setup be run over a modified ethernet cable (serial comms + power?)
I've just about got my imaging setup with the dome totally automated and seeming to (mostly) get through the night, so this little project is starting to move to the forefront of my mind.
I priced up an AAG Cloudwatcher from Sidereal Trading, with the relative humidity and wind sensor, and that's going to cost just over $1000, and based on what all the parts cost for your setup (including an annemometer) is around $200.
The obvious advantage is the AAG setup is plug and play though!
My device is actually located within my obs so the rain sensor is not normally exposed to the elements unless the roof is open. The point of it was to alert me if the roof was open and it started to rain. At that stage I shut the roof and whatever got any rain drops on it would then need to dry out. I therefore don't really need a heater. It would be easy to add one and drive it from the Uno PWM pins if you needed it. I'm not trying to do remote imaging or a high level of automation - just seeking feedback of what's going on while I'm sitting inside the house.
In terms of cables, I've got USB and 12V power going to my enclosure. The power is only required for the anemometer and could run direct to that device.
Regarding the AAG, I think it uses a similar approach to the Boltwood in terms of ambient-sky temp to predict cloud. I think that gives quite a lot of false positives which is ok if you want to be sure however for me that would rule out a lot of nights I'd rather be capturing something. I'd also rather spend the $800 on something else. If all you need is the cloud monitor, this can be done with just the Uno and the MLX sensor - probably less than $50 in parts. I've added other sensors because I could and part of the exercise was learning to program the Arduino and build the GUI. If you really want wind speed sensing, with the benefit of hindsight, I'd probably consider one of these hot wire options as they are a lot cheaper and easier to incorporate in the project:
If it had some kind of cloud sensor, it would be perfect. You'd just need to have a listener for the data (apparently can run off the WiFi and you can read UDP packets or Bluetooth LE). Kind of expensive, but still cheaper than the AAG system (minus the cloud sensor)
I particularly enjoy the rain sensor using haptic feedback. Kind of novel approach.
If it had some kind of cloud sensor, it would be perfect. You'd just need to have a listener for the data (apparently can run off the WiFi and you can read UDP packets or Bluetooth LE). Kind of expensive, but still cheaper than the AAG system (minus the cloud sensor)
I particularly enjoy the rain sensor using haptic feedback. Kind of novel approach.
I'm a little bit shy on the crowdfunding gear - I waited for 2 years on one thing, and it eventually arrived, but by then, things moved on and it was a bit irrelevant.
I am still waiting on something else, which is past 2 years now. I'm working to the theory that if it turns out to be a good product, I'll pay the extra and get them when they are manufacturing normally.
Maybe I just get back to basics and build something that tells me if it's raining, or cloudy or too windy so I can close the dome!!!
I'm a little bit shy on the crowdfunding gear - I waited for 2 years on one thing, and it eventually arrived, but by then, things moved on and it was a bit irrelevant.
I am still waiting on something else, which is past 2 years now. I'm working to the theory that if it turns out to be a good product, I'll pay the extra and get them when they are manufacturing normally.
Maybe I just get back to basics and build something that tells me if it's raining, or cloudy or too windy so I can close the dome!!!
It's been a while since I last updated on this project. I've been collecting additional data points to calibrate the model and now have 267 points including 91 with clear skies. With the weather turning cooler and more clear nights I should be able to get some more data points to improve the model as most of the data so far has been over summer. The clear sky emissivity model is showing some bias that should be improved with some more cold night data points. The last chart below shows the observed vs predicted cloud cover (0=clear, 10=cloudy).
Good update - looking forward to more in the future.
In the interim, someone on another forum pointed me at this rain sensor: https://rainsensors.com/
Just putting this here are as it fills out options for those looking to DIY.
It has different modes of detection, but it's as simple as a relay you can wire up which could be used to just trigger the dome to close directly, or could be read in to the arduino (it has a tipping bucket emulation).
Good update - looking forward to more in the future.
In the interim, someone on another forum pointed me at this rain sensor: https://rainsensors.com/
Just putting this here are as it fills out options for those looking to DIY.
It has different modes of detection, but it's as simple as a relay you can wire up which could be used to just trigger the dome to close directly, or could be read in to the arduino (it has a tipping bucket emulation).
I use that RG-11 rain sensor on my Nexdome observatory and can confirm it works well. A tiny drop on the detection dome and the Shutter on the Nexdome closes immediately. I purchashed mine from here https://oceancontrols.com.au/hys-001.html
I use that RG-11 rain sensor on my Nexdome observatory and can confirm it works well. A tiny drop on the detection dome and the Shutter on the Nexdome closes immediately. I purchashed mine from here https://oceancontrols.com.au/hys-001.html
Jeremy
That's really good to know - how do you have it hooked up?
Directly to the dome or through some kind of controller?
I'm pretty sure my Dome has hardwired switches that I can plug in to (which I've never bothered to look in to), and they appear to be exposed in the dome driver, so maybe that would work too - no other bits in the middle.
Ah, that's good.
I just sat through the manual of my dome, unfortunately the switches are OUTPUT switches, not input, so I'll have to revert back to a controller plan.
I've finally built my own version of this, pending one chip which I'm picking up tomorrow. Other than that, I'm getting readings from everything, which is encouraging!
I was hoping if you could share your code (the arduino sketch for formatting) and the PC project also?
Outside of the rain sensor, which obviously NEEDS to get wet, are the other sensors OK to be behind a clear cover (ie would the IR sensor and the light sensor suffer)? I'm trying to minimise holes in the case if I can. This will somewhat change the type of project box I select.
Thanks for putting together this project. It's been fun so far.
Hi Rob. I'm not sure about the IR sensor as it depends if the clear cover attenuates IR.
You can test the rain sensor by placing a drop of water on it with your finger.
You should be able to try it with/without the cover in place and see if the readings change. Happy to share the code - just need to work out how to get it to you.
I've purchased (on order) a couple of different project boxes, one with a clear cover. When it arrives I'll test the difference between beneath and above the cover and see what happens.
If you can use Dropbox (free for their basic version), it's pretty easy to transfer a file/zip etc. If you don't use it, send me a PM and I'll send you a referral link - it's simple to use.
I've got all the readings, but I'm trying to work on the best approach to get the data stored away and formatted. I'm going to probably do with the PC solution anyway (the observatory PC is always going to be on if I'm imaging), but I might make a version that could run independently on a Raspberry Pi or similar once I got a feel for the file creation.
