I want (need?) a wind generator.
Actually, I already have three but they're small.
Okay, where do I start? It seems looking at this site, which I've found most valuable, that there are a few possible starting points. The air rotor (I still like the word rotor, but you may prefer blades), the alternator, the system or needs? The optimum choice may depend on what you have to start with. In my case, it seemed most logical to start with what my location has to offer. Fortunately, NOAA has a monitoring site not far away. I understand that there are similar resources for many areas.
NOAA determined that our mean annual wind speed, a six-year average, is 10.3 mph. Not great wind, but enough to work with. Calculating the Rayleigh time distribution of wind speed for this mean produced some surprises. For example, I should have winds of 8 mph for more of the time, during a year, than any other speed. Interesting. Initially, I thought this meant that I needed a generator that was optimized for this speed. Wrong! I forgot about the V^3 power factor. Calculating the energy distribution for this time distribution shows that my maximum energy peak is at 16 mph. Ah, that sounds better, but this is still far off from the commercial machines that produce rated power at 25 mph or more. Obviously, I need to roll my own.
Another factor that impacts my starting point is that I already have some components I would like to use if I can. I have a set of blades (actually two sets of three blades) that are designed for 42% efficiency at a TSR of 4.5. It's hard to do much better than that. These are constructed of a tubular aluminum spar and formed aluminum skin over ribs using a high L/D (100), under-cambered airfoil (Wortmann FX 60-126). The blades have a constant 13" chord with no twist. The swept area is 201 ft^2, which is 16 ft diameter. I also have a variable pitch hub for these blades for which I intend to use active pitch control. I can already hear the groans in response to using aluminum and variable pitch. The consensus on this site seems to be that aluminum will fail from metal fatigue and variable pitch is too complicated to be practical. I have to wonder if these are the real reasons that they're rejected. In any case, we shall see. This will be a down-wind machine hence, no tail feathers.
Here's a chart showing the Rayleigh time distribution, the energy distribution, the available power from my air rotor and its RPM vs. wind speed.
Now we come to need. Most are going to say, "as much as I can get!" Sure, I feel the same way but also understand there's a practical limit. First, I want to stay with a 12 volt system. Why? Most of my off-grid loads, by far, are 12 V. I have no 24 or 48 volt loads and my two 12 volt inverters will handle the AC loads just fine. Yes, I understand that the higher current will incur more losses but I have a few techniques in mind to combat that. You see, I don't mind complexity. In fact, I welcome it - it offsets boredom. The KISS principle certainly has merit, but what do I have to give up to get it? I don't live off the grid so I don't have to depend on RE. Then, why do I do it at all? Simple - it's fun! Maybe someday I'll save a few bucks (problematical), but if I don't, I can still say I've enjoyed myself. However, with increasing energy costs, it may pay off sooner than expected. Our grid rates are now up to 26 cents/kWh. The number I picked out of the air (ha!) for the wind generator is 1 kW at 12 V at 16 mph wind speed - a low speed machine.
Okay, so what's next? Guess I better take a look at the alternator. The axial flux machine was unfamiliar to me before coming to this site. What a neat idea! It looks like it should be ideally suited for what I want - simple, low speed capable and good output - until I saw efficiency numbers of 50% being bantered about. I've got this thing about efficiency - small numbers don't get my attention. So, what's the deal? Is 50% all I can look forward to? After digging out the formulas and cranking them through a spreadsheet, I see that it can do better than that. It's the usual problem of trying to match up the components of a system - it takes a systems viewpoint. One thing I decided early on is that I'm not very much concerned about winds over 24 mph - they simply don't occur very often here. Also, a variable pitch air rotor can cure a lot of ills that fixed pitch can't cope with.
The variable pitch hub has a back plate of ¼" mild steel - looks like I can use it as one magnet rotor of a dual rotor alternator. It's 18" in diameter and heavily reinforced so distortion should not be a problem. Looking at the most bang-for-the-buck deal on magnets quickly showed that the 2x1x.5 magnets had it. At four bucks each for N40s from eBay, they have the advantage compared on unit volume. Unfortunately, a single magnet, of this size, per pole, isn't going to be enough. However, two in line radially, giving 4x1x.5 looks reasonable on an 18" disk with enough magnet to get the power in the RPM range (55 to 189) available. 16 poles will fit with reasonable spacing and requires a total of 64 of the 2x1x.5 magnets for the two rotors. This, of course, means 12 coils for 3-phase.
The numbers say that I need 32 turns for cut-in (13Vdc) at 7 mph (55 RPM, star (wye) connected) and a graphical analysis says that I can fit 3-in-hand #14 AWG (#9 equiv.) into the available space that yields 27 milliohms resistance per coil. Coils will be in a racetrack shape (parallel legs), not trapezoidal, and have an aspect ratio (outside) of about 2:1. This estimates a resistance of 0.216 ohm per phase (8 series coils). Rectifier losses are ignored at this point since I intend to use synchronous rectification.
Power output? Again, the numbers show 987 watts into the batteries at 16 mph (126 RPM), pretty close to the target goal of 1 kW. They also show a stator dissipation of 415 watts for an alternator efficiency of 70%. That's better! The power available from the air rotor is 1770 watts so it's nowhere close to stall. At the top end it's not so pretty. At 24 mph (189 RPM) output power would be 1872 watts but with a stator dissipation of about 1500 which yields an efficiency of 56%. Not too cool (literally!). Power available from the air rotor is 6 kW so again, far from stall. I think I'll set an arbitrary limit of 100 amps output, about 1300 watts, which will occur at about 19 mph and will limit the stator dissipation to 850 watts. Current will be limited by a PWM load control. This control will be active between 19 and 24 mph and at 24 mph the blades will be feathered and the machine shut down. Limiting the speed to less than 200 RPM should save on the wear and tear with hopefully, longer maintenance intervals.
What about controls? As indicated, the air rotor has active variable pitch. With a fairly simple controller, it can maintain a constant optimum TSR of 4.5 over the wind speed range of interest plus give shut down capability. Why 4.5 rather than the general consensus of 7 or so? My research shows that a modest solidity three blade rotor peaks in efficiency at 4.5. Lower solidity, high-speed rotors (especially two blades) do better at a higher TSR but the curve peak is narrower. I question the wisdom of using fixed pitch three blade rotors at high TSR. I guess I'm efficiency driven - please bear with me. The combination of pitch control and PWM load control provides something similar to what the utility-class machines are doing. As I said, I don't mind added complexity if it produces meaningful results, particularly in efficiency.
So there you have it. Of course, except for the air rotor, it's only a paper pipe dream. Any comments that may help me put it into the real world successfully will be greatly appreciated.
