Arduino Weather Station

Hello all, first time posting here. I have had an Arduino Uno since December of last year and it's a really neat product. I have played around with LEDS and a few sensors here and there just kind of learning the ropes of programming and such. However, I haven't really done a real project from design to completion yet.

So, I figured I would start with something feasible in my eyes. I want to build a small weather station using my Arduino Uno. I'm sure something like this has been done before, so a little lack in originality on my part, but the main goal of this is to advance my knowledge of Arduino hardware and software a little more.

I've done some thinking and searching, and I want the whole system to be stand alone, including the power source for the Arduino and sensors. I have a small solar panel I took out of a battery powered outdoor lamp. It used the solar panel to detect when the sun was down and turn on the light, as well as charge the batteries throughout the day. The circuitry inside had gotten wet and corroded, so I couldn't look at it to figure out how exactly it worked. I did check the solar panel's voltage output in bright sun and it outputs about 7.3 VDC and stays very steady.

However, doing some searching, I found this link: http://www.arduino.cc/cgi-bin/yabb2/YaBB.pl?num=1252704060
The link vxir posted to his personal site has a nifty drawing of a circuit that should work to accomplish my needs of detecting the light and powering on the Arduino.

So, basically, I just thought I would make this topic to share and get some input from other Arduino devs if this sounds like a good project and that circuit would work for my power/light sensing needs?

Cheers!

BUilding a weather station is a very good project as you can build it incrementally

  • outdoors temp
  • indoor temp
  • humidity
  • airpressure
  • lighting conditons
  • rain gauge
  • wind speed
  • wind direction
  • maybe more

I would not advise to experiment with the powersupply and the sensors simultaneously, make that separate experiments until you understand the quirks of the solar panel. Not that it will brake the Arduino but you will experience interference which can be quite annoying e.g. when try to debug sensor code you get a voltage drop :frowning:

succes!

Yea, I am currently looking at how to make a wind vane using a potentiometer. I figure the power supply system will be rather simple once I figure out my sensors, how much current draw they will have, if I need external sources for a sensor, etc.

I'm trying to build as much as I can from scratch, i.e building a wind vane instead of buying a pre-made one and interfacing it.

I will try to keep this thread posted when I make major achievements.

Have you checked www.instructables.com for ideas?

make a wind vane using a potentiometer

Two problems, most pots are quite stiff to turn so unless the vane is really large I think it will tend to stick.

Also pots have a dead band area which can be quite large, in that area you won't know there the pot is.

You could use two pots 180 degrees out of phase and swap inputs.


Rob

Wind vane... excellent.
I am also building a solar powered weather station. (well, started on it anyway) :smiley: I am using an Anarduino (eBay knockoff) outside, wirelessly send info to an indoor Arduino that compares outdoor temp to indoor temp, and pushes all data to an XML file. Eventually, I will have a PHP that queries weather.com, compares my data to theirs.

As for the wind vane, my design (so far only in my head) uses 3 or 4 hall effect sensors and a magnet on the wind vane. Maybe triangulate the hall effect data to determine position. Dunno.. works well in my head...

Also.. what do you think about using a motor with a fan on it to generate current to measure wind speed?

Also.. what do you think about using a motor with a fan on it to generate current to measure wind speed?

A stepper motor from an old floppy drive (remember those??) should work. Try searching for examples.

Don

So I've been working on the wind vane and I am running into some complications, mainly what some of you have brought up about using a pot, particularly the "deadspot" where it has no connection.

I snapped off the little piece of metal keeping it from turning 360 degrees, and the pot I have has a fairly easy to turn centerpiece. However, I'm using a voltage divider on the analog input and using the ADC to determine which direction it is turned. It works for the most part, but I'm running into trouble when it comes down to the final 90 degrees of rotation, basically where the "deadspot" is.

So, I might try another POT, or start relooking at my way of doing this wind vane using some of y'alls suggestions.

EDIT: So I started looking at Hall effect sensors, and this seems like possibly a more reasonable way to go, as there is less mechanical linkage (turning part of potentiometer could wear out over time) and can get a more accurate reading. So from my understanding of what I've read, I would need a Hall effect sensor at every point I wanted to define as a direction, right? Like if I wanted to keep it simple and just do North, East, South, West; I would need 4 Hall sensors then?

If you decide to use Hall sensors you could take a look at this post

My system has 8 Hall sensors and displays 16 wind directions. With 4 sensors if your magnet is wide enough to trigger two sensors when it is between them you could get 8 directions.

Ah, thank you EmilyJane, that's what I was curious about. I will be ordering some of these Hall sensors to play around with. :slight_smile:

YOu can also use a photoresistor [PR] and a light, and put a (plastic) disk on your vane.

Make the disk so that the disk covers the PR 0% at 0 degrees, 25% at 90degrees, 50% at 180, 75% at 270. 100% at 360. In short at every angle there is another percentage covered. With some tweaking I think you can make a lookuptable that can easily detect the 16 major directions

So not a round edge but some sort spiral inwards. bit like - http://www.karbosguide.com/images/4c_045.gif - couldn't find a better picture

OK around the 0/360 degrees border you have some trouble (as it goes from 100% to 0 in only a few degrees (which might be interpreted as another value)
that artefact can be solved with a second PR and a hole in the disk near that 0/360 border.

If PR1 reads 50% you know it is 180 degrees IFF PR2 = dark and it is 360 degrees if PR2 sees the light

A variation is to use 6 PR's and put many holes in the disk. The pattern of the holes are bits and with 6 bits you can define 64 directions.