# Export Anemometer Data to Computer

1. Nov 24, 2009

### minger

Hi all. My little sister is finally getting through high school and is asking about things she could try for a science fair. The teacher has already shot down a few ideas, so she asked me for a more difficult project.

The city that she lives in and I grew up in is proposing building a wind farm. Personally, I am anti-wind energy, but that is a different topic. I proposed to her to do a feasibility study for wind power, based on the effective wind energy in our city compared to the national average.

This would involve taking wind data, ideally continually through the day, and graphing it. We could then adjust the readings for height based on some sort of boundary layer theory to get an effective wind speed at height.

Now, I've looked around for digital anemometers and they bitchin' expensive; upwards of $600. She's a high school student so she has like a$50 budget. I seen a "tutorial" on instructables that builds a home made anemometer from spare "cups" a couple R/C bearings and a bike speed sensor:
http://www.instructables.com/id/Digital-Anemometer-wind-meter/

My question is if anyone knows a way to export this data to computer for post processing. Perhaps there is simply a particular gauge that can export, or someone knows a method that might be useful.

2. Nov 24, 2009

### FredGarvin

Using that bike speedometer would be difficult since the source signal is probably so low you'd need amplification and probably some signal conditioning. This may be a toughie to do cheaply. Let me do some noodling on this as well.

3. Nov 24, 2009

### mgb_phys

i would probably build an egg-cup annemometer which turns a little DC motor and then measure the output voltage with a digital multimeter (DMM)
You can pick up cheap DMMs with a simple 3wire serial output that you can use to read the displayed voltage into a computer - ad so calculate the windspeed.
The cheap ones are slow, they only read the voltage once/second or so - but thats ideal for windspeed.

If you want to make the project cheaper but more complicated you can use the soundcard's microphone input as a voltmeter - but you will need some external electronics. Or these guys have been making simple data capture stuff for pcs for years http://www.picotech.com/educational.html
It used to be much cheaper/easier in the days of the BBC micro/Apple II/etc - you could build this stuff directly.

Last edited: Nov 24, 2009
4. Nov 24, 2009

### FredGarvin

The DC motor is a good idea. I couldn't get my head past thinking about processing a frequency signal. The DC would be a lot easier.

5. Nov 24, 2009

### Topher925

DC motor? Why make it so complicated? Just use a simple encoder (LED and a photoresistor/phototransistor) connected to a PIC. Something like a 14pin PICaxe can handle reading the signal from the encoder and send that data to a PC over USB. All the required electronics should cost you less than 10 bucks plus maybe 25 for the programming cable.

http://www.sparkfun.com/commerce/product_info.php?products_id=8351

6. Nov 24, 2009

### Bob S

If you decide to use the bike speedometer design, put one magnet on each anemometer arm, and get 3 reed switch closures per revolution. Use a resistor and a capacitor to produce a pulse every time the reed switch opens (or closes), and use a leaky op-amp integrator circuit with about a 2 to 4 second discharge time constant, and perhaps 30 pulses per sec for full scale on the op-amp output. You can use a LM324 quad op-amp available at RadioShack. The output will be a voltage roughly proportional to the anemometer RPM.
Bob S

7. Nov 25, 2009

### minger

After posting yesterday, I got to thinking. I was considering simply using a ducted computer fan to put out a DC current and then just seeing if I can borrow a DAQ card from somewhere.

After seeing the replies here though, I think maybe I (we) are in over our heads. My electronics experience is minimal at best. Mechanically, desiging a system is no problem, but capturing the data seems tricky for me.

The multimeter idea is really good. The problem still seems to remain, though. How do I export and automatically save this data? I can help with postprocessing if I can just get it. Would I need to write some program to interface with the serial port and save as csv or simple unformatted text?

8. Nov 25, 2009

### FredGarvin

9. Nov 25, 2009

### mgb_phys

Yes, the meter just writes the digits from the display to the serial port every second or so.
You just need to read the string from the port as if it were coming from a file - there should be simple examples for most laguages.
And it's normally a very simple dumb port, no fancy handshaking or commands.

The DAQs from someone like PICO even come with drivers to access the serial data from inside excel.

ps. You may need a USB-serial converter if your machine doesn't have any serial ports (if it less than 5years old) - eg http://dealextreme.com/details.dx/sku.24512

pps. Probably wouldn't use a PC fan, either look at making an egg-cup manometer or a model aircraft propeller - you can mount it on the fan's motor, but the PC fan is probably too small for the wind

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?