Wind Data

[wil]

I was just thinking about my wind data, that I record on site.

I don't know if this was talked about before or not on this board (I haven't performed any searches on the subject.) But I thought I would share my musings.

My Weather station says my average windspeed, so far, for the month is 6 mph. With the high being 49 mph.

I have the software record the current windspeed and last wind gust every 10 minutes.

I decided that I would take a look at the occurrences of wind which is 7 mph or greater.

The data is stored in CSV format so it's easy to perform analysis with spread sheet software.

Here is a look at a few days for this month:

Day  Speed(7+)  Gusts(7+)

(14       1             28)

(15       11            56)

(16       54            123)

(17       68            88)

(18       38            72)

legend:

Day - Day of the week.

Speed(7+) - Number of occurrences of wind at 7 mph or greater.

Gusts(7+) - Number of occurrences of wind gusts at 7 mph or greater.

This looks much better in terms of power in the wind verses the over-all average of windspeed. If you remember from above the wx software is saying 6 mph average for the month.

Getting back to averages, the best day for the month was probably the 6th of this month (Not included above) with an average of 20mph and a high of 34mph. The lowest speed on the 6th was 10mph.

Wil

[Flux]

6 mph is very low.

You don't say how long your events are above 7 mph. The gusts may not contain significant energy.

If you go to Michael klemen's site you can find tables that estimate the power from average wind speed. This uses a Rayleigh distribution and takes care of the fact that speeds above the average carry proportionally more power.

The average of the cubes is not the same as the cube of the average.

With a 6 mph average for a month, you will get good power on certain days but the days with nothing will be high.

If this is for fun then ok but normally an average of less than 8 mph would indicate a site that will not do much good for serious wind power.

The effect of the height of your monitoring equipment compared with the intended height of the turbine could have a significant effect. Normal monitoring height for airports met office etc. is 10M (30ft) If you can get the turbine well above that then things will look better.

Flux

[asheets]

What kind of anemometer do you have, and what is the cup configuration?

[wil]

Flux,

Thanks for reminding me about Klemen's site. After seeing his article "Wind speed Data and its Application to Wind Generated Power", I remembered that I read this before and had a copy of it stashed somewhere.

This article was one of the "reads" that inspired me to purchase the weather station. (Oregon Scientific  WMR968) Not high-end data acquisition but figured it should be close enough to help make the decision to play with wind turbines.

I don't think I will go as far as Klemen but have decided to try lower data acquisition times.  I will have to find the user manual, I don't remember how I set the time frame.

"average of less than 8 mph"

Well according to Klemen, average data doesn't mean that much nor did my time frames of 10 minutes.

I'll see what the data looks like after I change the data acquisiton times closer to 1 minute. This should also give a better idea of the duration these speeds occur.

Wil

[wil]

I don't have the specifics at hand but it's an Oregon Scientific WMR968 Weather station.

Wil

[asheets]

I have a similar one -- I ended up disconnecting it and getting an NWS-spec unit instead.  Far more accurate and a much lower cut-in.

[wil]

I haven't really had time to look for the manual yet, but at the site where I purchased the wx station it has the following specs:

"The anemometer measures wind speed and direction. The anemometer operating range is up to 125 miles per hour with accuracy resolution of +/-0.4mph. Data transmission occurs every 14-seconds."

What is the difference between this one and the NWS-spec?

I did take a look at several different anemometers, some very high priced and some fairly low cost (including homebrewed units). I had no idea the cost of these units until I started looking for one...

Incidentally - I started working on a little software program to help analyze the data that I'm going to recieve from the weather station software. I didn't plan on it but my curiosity was sparked after re-reading Klemen's article.

Wil

[willib]

wil , i made one(an amemometer) out of a pc powersupply fan , i havnt calibrated it yet , but it outputs a digital squarewave proportional to its speed.

[coldspot]

Sounds like a very nice unit to me!

I wish my amemometer, (self made using hosts cups),

had better memory and data logging.

I've gotten two speed read-outs hooked to it for now using the standard bike speed-o computer and also rigged a very old exersize bike speed-o up to it using the same reed switch. The Exersize bike unit was rigged on the bike at about the same O.D. as the cups O.D. not like the bike computer that would be rigged at 24"-28" while the cups are only about 1.5" this made me think it might be more close to real readings and they seem to be within a mile or so of each other.

Darn thing flew off the little tower I have it on now out to my shed/shop, (10 miles from town), but I did manage to find the cups after only a few days and even found the 1/2" x 1/8" Noe magnet that came un-glued from the side of the cups about a week later. I re-designed the the shaft mount the bearing rides on and that shouldn't happen again. LOL



Anyway great post I'm in envy of your toy !

[wil]

I just took a look at the data file size. The old file size was 4.88KB. The 20th was the first full day recording wind speed and gusts every 60 seconds. The file size is now 22.3KB.

I'll see if the software allows less than 1 minute; maybe 30seconds, then take another look at the file size. I don't want to let the size get too far out-of-hand since it's going to take months to gather data, and I don't want to have to baby-sit the files that could easily eat-up hard drive space.

I suppose what I could do is, create a script that compresses and archives the data. Then it would just be a matter of removing the archived data files every 2 or three months depending on size.

Wil

[asheets]

I'll grab my instrumentation book and get back with you tonight or tomorrow.  But I will tell you this -- NWS spec calls for an aerovane or a pressure plate for wind speeds greater than 70 kts, because of (1) the anemometer "curve" becomes unstable in that range, and (2) cup anemometers tend to fail at that range.

There's also specification for cup size, shape, radial distance, etc., that the Oregon Scientific ones don't match.

[asheets]

OK, here's the poop from "Instruments for Physical Environmental Measurements, Volume 1, 2nd Edition" by Wang and Felton, 1983, pp160-170 (this was my instrumentation textbook in 1991, so things don't change all that much).

The specs are:

*3 cup, for uniform torque balanced with a minimum of resistance

*semi-conical cup instead of hemispherical cup for rigidity

*beaded, rather than plain, cup edges to reduce turbulence effects

*a minimum of size and mass

*a linear relationship between tangental rotation and windspeed throughout the measurement domain, even if the domain is limited as a result.

Depending on the measurement domain, 2 different configurations are noted:

*7.5cm arm length with 5cm cups, for cutin speed of .5kts -- largely for research

*45cm arm length with 10cm cup with cutin speed of 5kts (domain limited to 4.5 to 47kts) -- NWS standard model manufactured by either Climet Instruments of Belfort Instruments

[wil]

Wow!..I kinda doubt mine has all that; I will have to look but I'll bet it falls short on most of those standards.

I'm certain it has 3 cups, I don't remeber if they are hemispherical or silightly conical, and I'm resonably certain the cup edges are not beaded.

Its been running for about 2 years now. I installed it when I starting thinking about a wind turbine. I took measurements for almost a year when I decided I had enough wind to start the project.

The smallest increment of time I can use to record data is 1 minute, so I will not have a narrow degree of accruacy for wind data, to find the power in the wind on my site. But it should get me close enough.

Thanks,

Wil