Arduino rainmeter

[CitadelCore]

Hey!

I'm thinking about building an electronic rainmeter from an Arduino and a project box. My idea is to have three elements in the system. The first is the controller, housed in a waterproof outside project box. The second element is a large funnel positioned in my garden, collecting rain when it falls. The third is a hose going into a container with a water monitor, and a solenoid controlling the water flow out of the container, which will go into our rainwater tank. I'm hoping that it can be connected wirelessly or wired to our home network, and be able to log the rainfall average over days, weeks, months, and years.

Only a few problems. How would I monitor the amount of rainwater that's currently in the tank? And are there any solenoids compatible with the Arduino? I'll be using a Freetronics EtherTen or a Freetronics Eleven. 

Thanks,
CitadelCore

[Kostas]

Google tilting bucket rain meter to get an idea of how it works. It should be fairly easy to make one yourself. You will need a reed switch that will give a pulse every time the bucket tilts, which means it has gathered a specific amount of water. The pulse can be fed to a 74XX series counter, which can keep count of the rain, not necessarily to an Arduino.

[ljwinkler]

Any solenoids are compatible with Arduino.
Just use a transistor and a diode.

Something like this will do:  http://jason-webb.info/wp-content/uploads/2011/09/Solenoid-Schematic.png

Cheers,
LJ

[dannyf]

"How would I monitor the amount of rainwater that's currently in the tank? "

Depending on the sensor you use.

I capacitive sensor would be my choice here.

[Sal Ammoniac]

How big is your tank? I don't collect rainwater, but I have two 5000 gallon tanks that I pump well water into. I use an ultrasonic sensor mounted at the top of the tanks to range the distance to the top surface of the water and use that info to calculate the amount of water in the tanks. These tanks are about six feet in diameter and about ten feet tall.

The sensor I use is this one: http://www.maxbotix.com/Ultrasonic_Sensors/MB7589.htm, which is designed for use in harsh environments like a water tank (which can reach 150F and 100% humidity in the summer).

[retrolefty]

We used just about every possible level instrument at a refinery. The one demanding the most accuracy and repeatably was a large vat that was used to batch process in three different liquids of different specific gravity. For that application they used a small PLC type system utilizing strain gauges that measured the weight of the whole vessel. Worked great but was not cheap.

[AndreasF]

Since you have to dump the collected water regularly, you might be able to measure the time it takes to drain the tank through a small pipe or tube. The flowing water could be used to make an electric connection between two contacts.  This likely wont be a linear relationship, but probably still fairly predictable.

[xtoffer]

"How would I monitor the amount of rainwater that's currently in the tank? "

Depending on the sensor you use.

I capacitive sensor would be my choice here.

I have experimented a bit with capacitive moisture sensing. It's fortunately very easy to build one yourself for rough values and it has the advantage that you can adjust it to the size you need.

You'd need two coplanar plates (placed close to each other) running the length you want to sense and someway to isolate them from the water (I have experimented with copper tape). The capacitance will change depending on the amount of surrounding water.

Then you only need somehow to measure the change in capacitance. It will probably be rather small in the pF region. There are different ways but I found timing a charge-up of the capacitor through a rather big resistance works well and requires very few parts. Since the time it takes to charge it to some value (some kind of time constant) is proportional to the capacitance (t=RC) this can be used as a sensed value. Unless you charge it to the right value it won't be the official time constant but it doesn't really matter as long as you get a value that is dynamic enough.

Issues: It is sensitive to other stuff than water that happens to come close as well. So if a bird happens to cozy-up to it, it might read wrongly (in most cases as in more water).

[dannyf]

I use copper tapes too: I paint them with polyurethane paint, and then paint the painted tape onto the inside of two plastic plates (with spacers). So rust is never a problem.

[dannyf]

The reading can be done through various ways of reading capacitance. Easier on the PIC but an outboard rc oscillator is simple to build and measuring its output frequency is simple.

[macboy]

Google tilting bucket rain meter to get an idea of how it works. It should be fairly easy to make one yourself. You will need a reed switch that will give a pulse every time the bucket tilts, which means it has gathered a specific amount of water. The pulse can be fed to a 74XX series counter, which can keep count of the rain, not necessarily to an Arduino.

+1
Tilting bucket is the way to go. It is elegant, small, and relatively maintenance free. And there is no bucket of stale water to turn into a disgusting stew of bugs and bacteria after you don't empty it.

[TheDirty]

This thread discussed water level measurement.  Someone posted a link to an article showing how to implement the capacitive method.  I'll be trying this out soon when the parts arrive for it.

https://www.eevblog.com/forum/projects/absolute-pressure-sensor/

[xtoffer]

This thread discussed water level measurement.  Someone posted a link to an article showing how to implement the capacitive method.  I'll be trying this out soon when the parts arrive for it.

https://www.eevblog.com/forum/projects/absolute-pressure-sensor/

If you are referring to the linked pdf it is -one- way of doing it. Interesting with the extra tube on the side though, curious if that improves stability somehow to just have it on the side of the bucket. The reading implementation can also be done in a number of ways, frequency through a lowpass with capacitor, charge timing of capacitor, capacitor dependent oscillator..

In that thread, pressure sensing is also heavily discussed, to me it feels like somewhat of an overkill (and not very cheap) for a small bucket measurement but that must be up to the OP to judge I still think that the cheapest/simplest would be the isolated plates with the charging measurement type.... but then again, what fun is taking the cheapest alternative :p

Oh and the tilting thing looks very interesting and, assuming you can balance it, also rather simple and cheap.