The Windmill Dream

[wind4Reg]

I have always been fascinated by windpower and have always wanted to have my own windmill. I am one of those crazy VAWT enthusiasts who wants to prove that a VAWT can indeed work well enough to supply my needs. I am the type of person however that would not be happy with a small battery charging system. I want a real system, one that can supply 95% - 100% of my electricity needs. My average consumption is about 1200KWH per month = ~14400KWH/year. Working with this number I would need a windmill in the 15KW range so that is what I am setting out to accomplish. Currently the electric company is not too keen on grid tied systems, so I would have to build a grid tied system that worked with the grid but does not supply power into the grid. This would allow me to use the grid when needed instead of maintaining a battery pack.  

First I built a small prototype of a drag type version that I hoped might have some lift assistance. Here is the pic of that original mill:





I mounted this mill on the roof of my garage. The rotor was sitting at ~ 28' high.

Unfortunately this mill performed badly. It turned very slowly and it soon became apparent that I would have to try another design. This time I went with a H rotor design. I used the NACA 0018 airfoil and made the blades 36" long and made the rotor diameter 60". Here is the pic of that mill:





This mill worked really well, it selfstarted and turned quite fast. I still have to get some monitering tools like a digital RPM meter so that I can determine the speed of the mill.

Next I needed to get some idea of the size of mill I would need to be able to produce my desired 15KW. I looked at other H rotor manufacturers to see what size thier rotors were for outputs in the 15KW range. I checked Solwind and the closest model they have is a 10KW H rotor that has 6 meter blades and a 6.3 meter rotor diameter. That's a huge machine. Then I checked "àç-ëïîôáëô" firm who builds a 20KW H rotor and they use 5 meter blades with a rotor diameter of 7.2 meters. So it looks to me like my mill will have to be in the 5 - 7 meter range for blade length and rotor diameter. I did some initial calculations and if my math is correct it looks like I would need a swept area of ~ 56.7 m^2. Knowing this, if I settle on making the blades 6m then my rotor diameter would have to be ~ 9m  for a rated power of 15KW at a windspeed of 10m/s and assuming gearbox and bearing efficency of about 90% and a generator efficiency of about 80%.

9m is too big, so I have to increase the efficencies and consider the tower height for more power from a smaller unit. I have settled on a tower height of 24m (80') which will put my the bottom of the blades above the treeline by a few meters. When I perform the same calculation on the 20KW mill above with the 5m blades and 7.2 meter diameter, it puts it in the 10KW range at 10m/s winds. So they must be rating their mill at 12-13m/s to be able to claim a 20KW rating. I then took a look at the Neuhaeuser 10KW H rotor. This mill has 5.6m long blades and a rotor diameter of 6m. So this is fairly consistant with the Solwind size for a 10KW unit.

So for initial sizing estimates I have decided that I will go with 6m blades and ~6m  rotor diameter. At a hieght of 24m, it should produce at least 10KW in a 10m/s wind and much more in higher wind conditions. In low wind conditions of around 4m/s it should put out over 1KW and in moderate winds of about 7m/s it should put out about 5KW. This project will need a significant amount of engineering to ensure this unit is safe.

Now that I have some idea of the mill I will need to construct, I can go back to my little model H Rotor that I already build and start experimenting with it some more.

We had some very windy conditions for almost 2 weeks straight and the model mill held together. I had used some light metal brackets to attach the blades to the cross arm and these brackets were showing signs of metal fatigue so I took the mill down. I have since replaced the brackets with heavier metal brackets and welded them to the cross arm where before they were bolted. I also got a pop rivot gun and pop rivoted the trailing edges of the blades since I had used hockey tape previously and it was failing. Then I ran wires from the top and bottom of each blade to the central area of the cross arm to provide additional support for the blades.

I purchased a 108VDC motor for $10 at Princess Auto and am currently in the process of connecting this to a chain drive on the mill. I can see already that this is not going to be a very efficient power transmission to the DC motor, but with limited resources this will have to do for now. I should have the unit back on the garage within the next few days. I am not expecting any significant voltage from the DC motor since I already tested it in my drill press and it needs very high RPMs to get the rated voltage of 108VDC. Plus my chain drive is a 30:7 ratio so I am not going to get very high RPMs on the motor. At least the motor can provide me with some estimates on the rotor speed since I catalogged the results at 16 different speeds with my drillpress.

Right now I am trying to figure out what alternator, gearing, and electronic controls will be needed for my big mill. I also have an interesting idea for overspeed protection on my big mill that is very cool, it is a magnetic disc braking system.

Now to get that model mill back on the roof of my garage and get some data.    

[MountainMan]

Reg,

Looks like you are having lots of fun with these plans.  Sorry to be a wet sock, but  you are headed down a dangerous path.  I put up a diary suggesting something along these lines but a little bit smaller just last week.  The experts around here quickly set about convincing me that it is too dangerous.  The centrifugal forces on the wings on an H style mill of large size are really amazing.  People have died from flying parts when such experiments have flown to pieces at high speeds.  A Lenz or Savoniuous style VAWT spins a lot slower and thus is much safer.  I've decided to experiment with scaling up the Lenz design that Ed at WindstuffNow is experimenting with.

Be careful!

jp

[bob g]

just a couple of thoughts, observations, and ideas

i think we all have had illusions of a large machine designed to handle our needs as they are, i know i did, and i assume alot of other folks start down this path in the beginning.

unless you have alot of time, money, equipment, and an understanding underwriter i wouldnt go the 15 kwatt unit of any design, be it vertical or horizonal.

the stresses and other problems in this class of machine are beyond the scope of most mortals. The damage and destruction can be very significant on a large machine and likely go up with the square of rotor diameter just as power goes up, and with the cube as when the wind speed doubles.

i suspect there is a formulae that correlates damage, and destruction and ties in nicely with the other wind formulae, but i digress, i hope you get my picture.

so what is one to do, you my ask?

first off do a carefull analysis of your power needs, cut back and conserve everywhere possible, much easier to conserve than to generate, and far safer.

second consider using multiple smaller machines to get to the power level you need. you not only have redundancy, but smaller generally is safer, easier to build, maintain and manage.

about 8 years ago i too thought about an H rotor machine in the 12 to 15 kwatt range, but luckily i didnt have the land yet and had to do more thinking and planning., i can now get by quite nicely with a projected windgen capacity of 500 -600 watts.

just my opinion, yours and others will likely differ

bob g

[rotornuts]

I liked your first model better. I think the reason it didn't work well is your blade setup.

 If you were to take the blades and turn them over so the flat panel is on the inside then you would have a blade that is producing lift towards the inside of the circumference. Both Ed and I seem to think that having the lift moment towards the inside is a better configuration than towards the outside. In fact if you flipped the blades and set the trailing edge 9 degrees to the inside of the circumference then you have a "Lenz Turbine". The 9 degree angle puts the lift moment ahead of the axis and was tested by Ed on his unit.

Try it it should work.

[electrondady1]

reg , you were on the cutting edge of design and didn't realise it ! flip that baby around like roto suggests and build a pmg for it . when thats done build another twice as big.

 if your buying at princess auto you must be here in canada. were you at?

[Ungrounded Lightning Rod]

Also:  If you make a lenz or savonius a bit bigger than a darrieus-style (half again as high, half agiain as wide, or about 23% bigger both high and wide) it should give you a tad more power than the best darrieus while still turning with the tips moving slower than the wind rather than several times faster.

You need more magnets and a bigger radius in your genny, or a higher ratio in your gearing/belt/whatever, due to the higher torque/lower speed tradeoff.  But IMHO the safety is worth it.  (Also, you've already got a really wide disk at right angles to the wind where you can build a REALLY large-radius axial-flux genny if you feel like it.)

Engineering a Darrius not to resonate, fatigue, and fail - with pieces flying outward at a couple hundred MPH - is HARD.  Several companies with professional engineers have tried and failed.  A savonius is garage-tinkerer rule-of-thumb stuff.

(And a Lenz probably will be, too, if Ed ever collects all the dimensions of a particular model in a single posting.  B-) )