I did a bit of research on how to best protect the steel parts from the elements. You can actually just leave the steel bare and let it rust. They say that it will not affect the performance in any way and will still last plenty long. Some people just spray the parts with a can of spray paint, and that prabably works fine too. Even better is to use these high quality industrial epoxy paints (like you find on street light poles), but I found from calling around that they were very toxic and expensive.

I think everyone would agree that the very best way to protect the steel parts is to have them powder coated, which is an electrically charged polyester powder that gets baked onto the steel at around 500 degrees. Anyway, I found a guy with rediculously good prices. He said he would powder coat 4 or 5 wind turbines for his minimum invoice amount of $100! That is half the price of doing the high end epoxy paint. So, I picked a textured green color and did both turines for this price...and he would have done like 5 or something for this price if I had that many! I went ahead and coated both turbines and they turned out great and look very professional. I am very pleased.

One thing to remember next time is to make sure to have the steel disks for the magnet rotors powder coated before placing the magnets and resin on them. You can't powder coat the finished magnet rotors because the resin would melt and the magnets would demagnetize from the high temperatures. So I decided to paint the magnet rotors using a steel primer and spray enamel paint. I think this will hold up fairly well for these parts.

I don't want to spend too much time writing about the math used to analyze wind turbines because it is not very interesting for most people to read and it has already been done a million times for these 10' turbines. I just think it will be interesting to see how these numbers will compare to the real world numbers that I get out of my turbines when they are finally flying.  I'm not really a big math guy so keep me honest here if you see any mistakes. 

Ok, it seems that the math for estimating the power output goes somehing like this. First you can figure out the total power available in the wind using the first formula below.

Pa = 0.5 x rho x A x V³

This is the power available to your wind turbine and is determined by the air density, size of the blades (swept area), and speed of the wind. Notice that increasing the swept area (A)and the wind speed (V³) create substantial increases in available power. You can't change the wind density so forget about that variable (rho). Doubling the diameter of the blades (which increases the swept area according to the formula πr² ), will increase the power output by a factor of 4. Also,doubling the wind speed will increase the power output by a factor of 8. So, you can greatly increase the power output by making the blades really big and getting the wind turbine as high as possible, since the wind increases substantially with height above ground.

All the indented stuff below is taken from a dicussion at http://www.awea.org/faq/windpower.html that was written by Eric Eggleston on February 5, 1998. He explains all of this by saying...

...

Power in the area swept by the wind turbine rotor:

Pa = 0.5 x rho x A x V³

Pa = power available in watts (746 watts = 1 hp) (1,000 watts = 1 kilowatt)
rho = air density (about 1.225 kg/m³ at sea level, less higher up; about 1 in Denver, Co)
A = rotor swept area, exposed to the wind (m²)
V = wind speed in meters/sec (20 mph = 9 m/s) (mph/2.24 = m/s)

This yields the power in a free flowing stream of wind. Of course, it is impossible to extract all the power from the wind because some flow must be maintained through the rotor (otherwise a brick wall would be a 100% efficient wind power extractor). So, we need to include some additional terms to get a practical equation for a wind turbine.

Wind Turbine Power:

P = 0.5 x rho x A x V³ x Cp x Ng x Nb

P = power in watts (746 watts = 1 hp) (1,000 watts = 1 kilowatt)
rho = air density (about 1.225 kg/m³ at sea level, less higher up)
A = rotor swept area, exposed to the wind (m²)
V = wind speed in meters/sec (20 mph = 9 m/s)
Cp = Coefficient of performance (.59 {Betz limit} is the maximum thoretically possible, .35 for a good design)
Ng = generator efficiency (50% for car alternator, 80% or possibly more for a permanent magnet generator or grid-connected induction generator)
Nb = gearbox/bearings efficiency (depends, could be as high as 95% if good)

...

Notice from the equations above (the Cp or Coefficient of performance variable) where he says the maximum power that a wind turbine can theoretically generate is ~59% of what is available in the wind. This value is called the Betz limit and was discovered in 1919. Above this value the wind appearently backs up in front of the blades or goes around them and just can't be converted into power. This is basically the efficiency limit of the wind turbine blades. Appearently, most well done homebrew wind turbines are doing very well if they to approach a 30% Cp, unlike some commercial turbines, which are getting much closer to the Betz limit (I'm not sure how close to 59% they are getting).

So from this discussion we are now able to figure out roughly what our power output will be if we build everything right on this wind turbine. It is important to point out though that you have to make sure to account for the loss of efficiency in the bearings and alternator, as well as the blades, which is where Cp, Ng, and Nb come from. From what I have read, it seems that the blades, bearings, and alternator all need to be very efficient to get even 30% overall efficiency and this is about where people say these 10' homebrew wind turbines tend to run. The numbers I have used in the formulas below give it about 30% overall efficiency for this reason. I don't really know exactly how efficient these individual values are in reality, but I think these are close. So the numbers for this wind turbine would look something like this...

P = power in watts
rho = 1.05 (density of air around denver, co)
A = 10 foot diameter (which equals 9.57 meters squared)
V = 10 miles per hour ( which is 4.47 meters per second)
Cp = 40% (so we have a very efficient rotor)
Ng = 80% (since we are using good quality bearings)
Nb = 95% (since we have a very efficient direct drive alternator)

P = 0.5 x rho x A x V³ x Cp x Ng x Nb
P = 1/2 x 1.05 x 9.57 x (4.47)³ x 0.40 x 0.80 X 0.95 = 136.41 (watts in 10 mile/hour wind)
P = 1/2 x 1.05 x 9.57 x (6.71)³ x 0.40 x 0.80 X 0.95 = 461.44 (watts in 15 mile/hour wind)
P = 1/2 x 1.05 x 9.57 x (8.94)³ x 0.40 x 0.80 X 0.95 = 1091.33 (watts in 20 mile/hour wind)

Another helpful equation is one that helps you calculate the RPMs of your wind turbine at a given wind speed. You must be able to calculate RPMs over a range of wind speeds in order to build the alternator appropriately, so this is very important. The tip speed ratio (TSR) in the equation below is simply the ratio of the speed that the tip of the wind turbine blade is travelling divided by the wind speed.

Revolutions / Minute (RPM) = V x TSR x 60 / (6.28 x R)

V = Wind speed (m/s)
TSR = Tip Speed Ratio
R = Radius of rotor (meters)

So, for the 10' turbines we will have...

V = Lets us 10 miles per hour, which is 4.47 meters per second
TSR = 6
R = We have 5 foot radius which is 1.52 meters

RPM = 4.47 x 6 x 60 / (6.28 x 1.52) = 168.58 (in 10 mile/hour wind)
RPM = 6.71 x 6 x 60 / (6.28 x 1.52) = 253.06 (in 15 mile/hour wind)
RPM = 8.94 x 6 x 60 / (6.28 x 1.52) = 337.16 (in 20 mile/hour wind)

Now you can compare these numbers to the real world power curve from the otherpower.com folks below. I beleive this graph was produced from a real world test of their alternator performance, probably mounting it to a drill press or something.  "Alternator 1" below is the same alternator that I have built for these two turbines (using 2x1x1/2 inch magnets, 12" disks, and 140 turns of 17 gauge wire). Comparing the numbers we computed above, you can see that the numbers are a bit different but not too badly off. This approach would give us numbers within the relm of what the wind turbine will produce. I think the values drop off in this real world test primarily because the alternator becomes less efficient at higher RPMs. I guess you loose more power to friction in the bearings as well. Anyway, that is enough math...building the turbines is much more fun...

Well I finally have a tower stub put together. This is the part that the wind turbine (the yaw bearing actually) slides over at the very top of the tower. The wind turbine needs to pivot or yaw in order to point towards the wind. You may be surprised to find out that people don't often use ball bearings for this moving part. I assumed at first that pillow block bearing or something like that would be best, but not so. They mostly use this kind of tube over tube design with lots of axle greese and sometimes a bronze or plastic bushing between the steel tubes. It turns out this design will keep the metal parts lasting plenty long, and it is actually prefered because the added friction keeps the turbine from nervously jittering back and fourth to face the changing winds. This way the turbine appearently stays pointed in the direction of the prevailing winds and produces more power. I like this keep it simple stupid approach since it is cheap and extremely easy to build...sometimes it is better not to over engineer something simple that just works.

I decided to build my tower stub a bit different from the design in the Homebrew Wind Power book. Instead of simply sliding the yaw bearing over a 2 inch pipe with some grease, I decided to use a slightly smaller diameter tube so I can slide a bronze or plastic bushing down the entire length of the tower stub and place a bronze washer at the top.

Below you can see the 2 inch piece of tube I started with. The outer diameter is exactly 2 inches. Look how crazy thick the walls are on this tube!!! This is expensive tube that I found in the scrap yard for only 40 cents a pound. The yaw bearing would never wear through this even if there were no bushings. It took 15 minutes to make one cut of this tube with my new steel cut-off saw!!!!

Anyway, I welded it to a 1/4 inch thick sqaure plate that can be bolted or welded to the top of the tower.

Below is the steel cap I welded that slides over the top of the tower stub. You put this on after sliding a section of 2 inch bronze or plastic down the length of the tower stub. This keeps there from being any steel on steel contact down the vertical legnth of the yaw being.  I am going to order a bronze bushing tonight that should fit just right. I am also ordering a bronze thrust bushing that sits on top of the steel cap and keeps there from being any steel on steel at the top of the yaw bearing.

Above you can see the tower stub with a section of PVC and the steel cap. PVC was my first idea, but since I found a website you can order bronze from I have decided to use it.

When we are ready to set this thing up we will grease the entire tower stub, add the bushing, and finally slide the entire alternator assembly over the top. That is all there is to it. The grease and bushings will allow the wind turbine to spin around and face the wind.

I just ordered the bushing from the website below...

Vertical Bushing

Horizontal Bushing (Fluid Bearing)