Wireless garden sunlight meter, feasible?

Hi,

Do you think this idea is feasible? I did a search on google to see if something similar has been done which I suspect would have turned something up if it was possible, but nothing.

My vision is to stick a bunch of sensors at various locations in my garden to record how much sunlight the various areas are receiving, log this information and have it wirelessly accessible via my laptop. Just typing that makes me think it is the stuff of science fiction, but then who would have thought I would see the day 3d printers were retailing at £2000.

My experience with electronics is close to zero, but I would say I am an intermediate coder, and a fast and enthusiastic leaner :slight_smile:

I could just spend the day in my garden and make notes on a piece of paper but it would be cool to get an idea of what happens over the course of the year, and also a project to satisfy my inner geek.

What do you reckon?

Learner*...I'm not an enthusiastic leaner :slight_smile:

That is not a problem. Take a Solar Garden Light, a ATtiny85 and a wireless link of your choice (RF433, NRF24L01+, etc). Remove the LED from the Solar Garden Light and use the Solar cell for both measuring the sunlight and charging the battery that will power the ATtiny. Send measurement reports over the wireless link to an Arduino that interfaces your PC/iPhone/Android/etc.

Cheers!

It certainly sounds feasible. How many and what sort of distance will these devices need to transmit the wireless data and how often do you want to take readings as these will effect the hardware needed and it's power requirements.

Awesome, thanks for the quick response guys.

I guess I will start with two an may go up to as many as 6 if it's working out. Max distance of 15 meters I reckon.

In general, the correct ratio of answers to questions is a minimum of 1.0. Higher is OK. Lower is not.

How much data? How often? How much are you willing to pay? How critical is data loss? Interference?

I suspect you'll find that a solar panel is not a very good way to measure sunlight intensity unless you also do a lot of work to control the load and measure the current/voltage characteristics of the panel. What might be more practical is to power the Arduino with a battery (try charging the battery with a solar panel if you're feeling keen) and then use an LDR to measure the sunlight intensity. They're available for different brightness ranges and there is plenty of scope to fine tune the behaviour by positioning/shading/filtering the sunlight.

Use Solar Radiation report site instead of build your own, It will give last few years data.

sonnyyu:
...

1.Solar Radiation (Watts/m2)

Solar Radiation, noaa.gov

go though history daily report, found out Solar Radiation data. use system total Watts with Solar Radiation data to define solar panel size and batteries.
...

The sensor might need match or calibrate base on sun wavelength spectrum.

sonnyyu:
...

...

China_Arduino:
...

My experience with electronics is close to zero, but I would say I am an intermediate coder, and a fast and enthusiastic learner :slight_smile:
...

Did you know hardware/electronics person also has kids to feed, Could you slow down? 8)

Just kidding, I am neither hardware nor software guy.

PaulS:
How much data?

As how much data do I expect to be transmitted? I really don't know but if it's just an integer representing the current temperature on an hourly basis I would not expect it's much. What do you think?

PaulS:
How often?

On an hourly basis would be sufficient I think. Maybe I could even configure it to run only during daylight hours, maybe 05:00 - 21:00.

PaulS:
How much are you willing to pay?

It would depend how well the solution is likely to work, if you think I can create something which comfortably meets my requirements then £150 - £200

PaulS:
How critical is data loss?

You mean if data is not reported for an hour or two. Not very important but it would be good if I knew if data was missing.

PaulS:
Interference?

I can't think of any reason why there should be interference in my location in particular.

The OP hasn't said what exactly he wants to measure - hours of sunshine or energy measured in watt-hrs per square metre.

The UK Met Office still uses a glass sphere that causes the sunlight to burn a track in a piece of paper to measure sunlight hours. I suspect it mustn't be easy to calibrate an electronic system to give the same numbers.

If I was measuring energy I think I would use a small solar panel with a known output in bright sunlight(unless there is a simple way to relate the resistance of an LDR to watts/sq metre). It should not be too difficult to program an Arduino to vary the load on the solar panel to keep roughly at the maximum power point and then measure current and voltage from time to time. The problem that will be faced is the huge and fast variations in energy levels if the OP is in an area that is affected by cloud cover. One would probably need to sample every 5 seconds. Of course the Arduino could integrate the samples and just report data once per hour or whatever is wanted.

...R

Interference refers to what happens if two devices try to send at the same time. If you have only a few devices, and data once an hour is often enough, interference is not likely. 10,000 devices trying to transmit every 30 seconds will result in a lot of interference.

There are several types of radios that you might use, ranging from 433 MHz transmitters and receivers up, in price, to XBees. The XBees take care of interference checking, retransmission, etc. As the price goes down, the amount of work you need to do goes up. The range typically goes down as the price goes down, too.

Robin2:
The OP hasn't said what exactly he wants to measure - hours of sunshine or energy measured in watt-hrs per square metre.

I am glad everybody here seems to know what they are talking about, the longer the thread gets I realise I don't have a clue.

What I need to know for example, area x in my garden I know faces the sun for y hours a day, but with the shocking start to the summer we are having in the UK right now it's probably averaging half that in reality with the bad weather. So I would say I am looking hours of sunshine, but surely energy measured in watt-hrs per square metre energy measured in watt-hrs per square metre could be used to get the same result? If a minimum level of energy is detected the sun is shining and return 1, if not return 0.

Am I barking up the wrong tree?

It is all do-able, but you will need to roll up your sleeves and start to learn how to do it. The first step is to break down the task into smaller sub-tasks, for example, measuring how much sunlight a single area gets. You will need to need to get a photo sensor, and then calibrate it to your garden, maybe when you read it, you will decide that a reading of at least 100 counts as being in the shade, and 500 is in full sunlight. At the first stage, you would not worry about more than one sensor, battery operation, or it being wireless. Just read the sensor, and write out cloudy/sun/etc. At this stage, you would use the delay function, and take a reading every hour, displaying it on the USB serial device and use the serial monitor on the IDE to keep track of it.

After that, you start thinking about multiple sensors. Initially, I would suggest just using wires to connect each of the sensors, but ultimately you have to thinking about your budget, etc. You would need to think about whether you want several sensors each wireless and with their own battery with a base station that reads each of the values (can be expensive, based on how many sensors you want, and the cost per sensor), or whether it is acceptable to have random wires strung about your garden. I tend to think you ultimately want each sensor to be an independent processor with sensor and battery.

You likely will need to think about perhaps protecting your sensors/microprocessors from rain.

If you are going the wireless route, after getting the sensors, you need to integrate radios, and batteries. Or instead of radios, perhaps use EEPROM during the day to record each hour (or more likely every 5-10 minutes) recording, and then you bring in the sensor at night, to read the values, and recharge the batteries. If you want the sensors to go longer, then you might think about putting a small micro-sd card on each sensor.

Bear in mind, given you are starting from scratch, it will be a lot of work. It might take longer than this season, but you would have it for the next season, and presumably by that time, you would be experienced, and add more things like maybe temp and rain sensors.

Remember Mr. Google and Mr. Bing are your friends, and can often be used to find other people who have done similar projects. For example, this came up when I used google to find weather station arduino: http://blog.makezine.com/2013/01/18/raspi-and-arduino-controlled-weather-station/

And this was found via looking for *arduino monitor sunlight*. http://blog.makezine.com/projects/make-18/garduino-geek-gardening/

A Seeeduino Stalker would be great for something like this. It combines the Arduino with a real time clock, SD card slot, and solar power management, and then includes a module interface for the wireless method of your choice -- including WiFi. They also sell a weatherproof case for it.

The big problem is that the price per unit is pretty high; it would be much cheaper if you could use one Arduino reading sunlight from multiple wired points in your garden, but you haven't really mentioned how big the garden is. Apparently it's quite large if the sunlight varies that much across it!

Thanks again or all the replies, this is honestly one of the most helpful forums I have posted on.

I am going take some time to do some reading and research but I will be back, most likely with more questions :slight_smile:

Just realized I forgot to include something in my earlier post - you really only need to measure the sun at one location in your garden (perhaps in your county!) - it will be the same everywhere within the degree of accuracy you are likely to be capable of. If you want to measure solar energy in areas that are in shadow (from your house or garage, say) then all you need is a clock to time when the sun gets to that spot.

...R

Robin2:
If you want to measure solar energy in areas that are in shadow (from your house or garage, say) then all you need is a clock to time when the sun gets to that spot.

Are you saying make a note of the time the sun hits that spot and then again when it's gone, subtract the two and Bobs your fathers brother?

Sort of ...

I mean that during the time the sun is shining in the area that was previously shaded the amount of energy falling in that area will be the same as in any other part of the garden. So the data from a single detector in an unshaded area can be used to figure out the energy in the shaded area.

...R

Robin2:
Sort of ...

I mean that during the time the sun is shining in the area that was previously shaded the amount of energy falling in that area will be the same as in any other part of the garden. So the data from a single detector in an unshaded area can be used to figure out the energy in the shaded area.

...R

Oh I see, gotcha.

Well I am putting this idea on hold for the moment, bought a mega 2560 yesterday, installed the IDE and been tinckering, great fun. There's lots to learn so will come back to this when I am more confident and more aware of the limitations and potential.

I mean that during the time the sun is shining in the area that was previously shaded the amount of energy falling in that area will be the same as in any other part of the garden. So the data from a single detector in an unshaded area can be used to figure out the energy in the shaded area.

Unless there are clouds around.