We now have “Climate Change” as a part of our everyday vocabulary, and this has spawned an interest in “green” power – power that is generated without burning fossil fuel. A small part of the energy needs of a country can be met by taking energy out of the wind. No doubt Wind Farms will continue to be set up at key points, but they will never replace coal fired power stations at our present rate of using electricity.
However, for a single house, wind power can become an important part of the mix in generating enough electricity to supply a family's needs. Only a few places will have wind every day, so for most of us it will be just a part of the solution. The obvious mainstay of any system will be solar panels, but these are still quite expensive, so I see the need for a cheap, easy to build, homemade wind generator.
Having lived the past 17 years without grid power, I have bought and built – and the wind has destroyed – many wind generators. Some looked good, some put out more power than others, but the common factor with them all is, that with in a few years they stopped working. In two cases it was my own fault in not foreseeing the consequences of my actions, but all the rest were ground to pieces by the force of the wind. However, the last generation of DIY wind generators look like working for the longest time - at least so far. I am passing on my experiences in the hope that it might save you a lot of time and a little money as well.
Up front, let me say that wind generators have the potential of doing great damage - either from the turning blades, or things falling off the tower. What I am writing about here are my experiences, and in no way implies that it will work for you. Do not take risks. If you do not have the skills, knowledge or experience, then seek out someone who does. You also need to check with the local authorities. If you try to build a wind generator, then you do so entirely at you own risk.
In this article I'll just talk about the actual wind generator, but to get useful power for a house you will need special batteries (deep cycle) and an inverter to change the DC volts into AC, and a "black box" to give over-voltage protection or regulation to the batteries. Some batteries are better than others, and certainly some inverters are better too. However, these two components of the system are outside the “home made” department, and we just have to buy the best we can afford. I have never bought new batteries, but I look forward to the day when I might be able to do just that. Some wet cell 2 volt batteries we are using are over 30 years old and have certainly passed their “use by” date, but just keep going. Gel cells rarely get passed 10-12 years.
If you are thinking of making your own wind generator, you will need to be able to weld - or have access to someone who can - and be able to purchase or build a tower to put it on. Some people use pipe masts that are multi guyed. Local councils may have restrictions on the height and position of any tower - so check first. I am also assuming that you are a handyman type and can come up with brackets, and put pulleys on shafts, and so on. Don't feel that you have to restrict yourself to what I put forward. As you get into it, incorporate your own ideas. Often you may have to improvise.
The basic components we need to build a wind generator are;
Blades – to be turned by the wind.
Hub – holds the bearings that allow the blades to rotate.
Generator – produces AC or DC current when turned.
Mount – allows the wind generator to turn into the wind.
What follows is a discussion from my experience about each of these components. I hope it saves you some time and frustration. If you would like more detail about a particular part of the process, please feel free to contact me via email.
While there is much information on the Internet about how to make blades from various materials, most approach the problem by designing something that will give maximum power at all wind speeds. Taking their advice, I have carved many blades from wood. However, I now think this is the wrong approach for those of us making our own wind generator. It works fine for the big boys that can control the speed of the blades, but what we need is maximum power at low wind speed, and maximum fouling at high wind speeds to reduce the maximum speed. Having blades that are designed for maximum speed in high winds puts too much pressure on the system. I feel the ideal is to have the wind generator putting out about 75% of its maximum power, in a moderate wind – lower stress equals a longer life.
While wooden blades will get you started – and they can be fun to make – they don't stand up to the harsh conditions all that well. In fact, one good hail storm with wind can finish them off. I tried putting aluminum on the leading edges, but that causes other problems if it gets loose. If you have to use wood, my advice is to use a hard and dense type of wood, keep the blades short to reduce tip speed, and have six or more blades so that they will foul the air at higher speeds.
At the moment, my preferred type of blade is made from black irrigation pipe. Anything from six to ten inch (150-250mm) in diameter is suitable. Obviously the bigger the pipe, the wider and stronger the blades end up. I find a length of about 700mm suitable, and this length can often be picked up as an off-cut. The black irrigation pipe has a couple of things going for it. It is UV resistant and just about indestructible. However, its real value lies in that the curve of the blade gives excellent starting power, and then the same curve causes air turbulence - and a braking effect - at higher speed.
When cut, the blades will bend slightly towards the inside of the pipe. Under pressure of the wind this is straightened out and doesn't seem to affect performance. If they bend too much then you either have the blades too narrow or the plastic is not the right type.
If I wanted to make six blades from a six inch (150mm) pipe, I would divide the pipe into three equal parts, and then cut lengthwise with a power saw. Then each third is again cut lengthwise, but this time on a slight diagonal. This gives a blade with less radius at the tip and more at the base. If you only want five blades, then just keep one as a spare. Another method of cutting blades I have used - and it works well for shorter blades - is to cut the pipe into six equal parts, and then cut off the outer trailing corner. I cut about one third of the blade's width, to about half way down the blade. Obviously, whatever measurement you decide on, then all should be cut the same. Once cut, the leading edges are planed (sharp hand plane) to a point from the inside to cut the air. The trailing edges can be just rounded off, or planed flat to the plane the blades will be moving in. Once the edges of the blades are smoothed off, you are ready to mount them on the hub.
To my mind, there can be no stronger hub than the front stub axle of a car or ute (pick-up). Cone bearings under slight pre-load will keep every thing rock steady. And they are just designed for turning. So far, all my hubs have been the older drum brake type, and I just run the V-belt on the outside of the drum. For the disk-brake type I think you would have to get a V groove machined into the edge of the disk brake, or come up with a way of attaching the outer part of a pulley. Obviously it makes things easier if you use the bolts that hold the wheel to be one of the bolts for the blades. It doesn't matter whether the brake drum has five or six bolts, but I have a slight preference for five (less chance for harmonics). You will have to drill additional holes towards the outside of the drum. Take out the brake shoes if the bolt heads foul, or grind the shoe down to fit. If you can activate the brake somehow, it could then be used to take the generator out of action in strong winds, but I have not tried it yet.
There is a knack in putting the blades on. The trailing edge needs to be flat as compared to the plane of the drum. For no particular reason, I make the blades to turn clockwise, so this means that if you are looking from the inside to the outside of the blade, the curved up bit goes to the right. I put a strip of steel between the hub and the blade to support the blade further out than the brake drum. It should be solid enough not to bend with the wind pressure. The steel strip is bolted to the hub with two bolts, one of which is the wheel stub. The blade is then attached to the strip with a further two bolts. The bottom of the blades should be just outside the drum to get maximum push from the wind. Keeping the 5 or 6 steel strips equal, drill two holes for the blade and two holes for the drum. Bolt the blades to the steel strips so that they are all even and the same.
Select one blade, make the trailing edge as close to an extended radius (at a right angle to the drum) as you can get it, drill the outer hole in the drum, and then do up both bolts. This first one then gives you a "point of reference" to space out the others. Attach the blades to the wheel stub bolts and tighten just firm. Each bottom of the blades should be the same distance from the center of the stub. Mark a common point on all the tips, or just use the trailing/leading edge at the tip, and space out the blades so that the distance between tips is the same for all blades. I find the quickest way to do this is to take a quick eye-ball guess, then accurately measure the distance between each blade tip and write them down. Add them up and divide by the number of gaps you need. Space out the blades using your calculated value and a tape measure, and drill the second hole for each blade through the drum and secure the steel strip.
Now it needs to be balanced. Wash out the cone bearings in petrol and dry, then add a little sewing machine oil. Put the hub back together, but only bring the center nut up to finger tight - no pre-load. You need the hub to turn with the minimum resistance. Mount the hub and blades vertically in a closed shed (clamp on to the top of a step ladder is one way). The least bit of breeze will upset the balance test. Turn the blades gently and see if they always stop in the same position - that is, find the blade that is heaviest. On the opposite side, add lead washers (squares of lead sheeting with a hole drilled through) under the steel washers of the bolts holding the blades on. If the bolts are long enough, leave the original nut in position and use a second nut. Keep testing and adding weights till there is no obvious "heavy blade". The "heavy" point may not be exactly opposite another blade, so split the counter weights between the two lightest blades. While getting the balance right is important, you don't have to split hairs over it. The greater mass of the brake drum seems to even things out.
The second way you can balance the blades - especially if the bearings are not allowing "free" rotation - is to have some wire or light rope coming out the top of a cone, and suspend the fan in a horizontal position. The heavy blades will drop below the horizontal, so then you can add weight to the opposite blades. The fan has to be supported exactly in the centre.
After balancing, take the hub apart and grease the bearings, and them do up the stub axle nut with about a quarter turn of pre-load. If you don't put some per-load on the bearings, they will flex and soon stretch the belt.
Let's try to keep it simple. For home made wind generators, the options fall into three groups. In the old days, the original choice was a DC generator.
These are identified by a commutator (a brass ring at one end made up of many segments) and carbon brushes. They require some maintenance to keep the brushes in good condition, but over all do a good job. The old Dunlite wind generators used this system. Problem is, they are hard to come by, and very expensive to have rewired if they burn out (as mine did).
The second option is a car alternator. They are relative cheap from car wreckers, they are almost indestructible, and if needed, the bearings and brushes are relative cheap to replace. You need to run them without the regulator. If you do not disable the regulator, as the batteries charge up in strong winds, the load drops off. This means the fan will go into overdrive and self destruct. Without the regulator, you can get 24 volts as well as the standard 12 volts, from an old car alternator. I have found that alternators from a small (4 cylinder) car work better than from bigger (6 cylinder) cars.
If you can get the speed up, a 12 volt alternator will quite happily put out 20+ amps at 28+ volts. I have run an alternator rated at 64 amps/14 volts from a small petrol engine with a 3:1 gearing up, and it puts out 25 amps without getting too hot. You need to remove/bypass the regulator, and use a solid (high wattage) variable resistor to reduce the 24 volts down to something like 7 volts (at high speed) for the field. My latest preferred method is a LM317 (with the required resistors and caps, see "Battery Powered Drills") going to the base of a 2N3055 transistor, and the output from that going into the base of four 2N3055's (all with 0.1 ohm resistors on the emitters) on a heat sink. The specs say otherwise, but my experience is that you need to keep 2N3055's under 1 amp for a long life. Up a tower, you would need some way of supplying 12 volts to the field when the wind was blowing - see paragraph below.
Do I hear you saying, "the perfect solution"? Not quite - there are still two problems. No matter how much wind you have, they will not produce output until there is power applied to the field (to activate the magnets). Once they are running you can feed power from the output back into the field to keep them going, but getting them started is the problem. Apart from fancy electronics to measure blade speed and then turn on the fields, the home-made type has two options. One, send a "blip" of 12 volt power up to the alternator field/s for a second every 60-120 seconds. This can be done with a simple to build electronic kit from Oatley Electronics. Two, have a permanent magnet motor/generator either connected to the wind generator, or in the same position with its own blades, that will supply volts to the field/s of the alternator/s when the wind is blowing. Once the alternator has started, extra current can be supplied to the field from its own output. I have used both, and slightly favour the small permanent magnet generator, because if there is no wind for days, you are not wasting power by sending a "blip" to the fields. However, the electronic "blip" saves another "turning thing", and thus has less maintenance.
[Update 2009. I read on the Net (but now can not find it again) where someone uses a flap counterweighted at the front, and a mercury switch set to switch on when the flap tilts back a bit. This sounds better than my other two methods.]
The second problem with car alternators is that they are not very efficient at low speeds. To get maximum output at low speed, the field needs about four amps at 12 volts. Thus, to get one amp of useful energy, you need to produce five amps of power. Of course the efficiency goes up as the speed goes up, but you are always losing about two to four amps to the field. In practice, four amps at start-up makes the alternator too hard to turn, so one or two amps is better for starting.
The third option, and by far the best in my opinion, are permanent magnet generators. They start outputting power as soon as they turn, and there is no power loss to the field. They can be bought, or can be home made - get good advice first, as a dozen permanent magnets can crush a man's hand. However, I think the best option is to recycle a Fisher and Paykel washing machine "smart drive" motor.
[Update 2011: The alternator as used in inverter generators uses permanent magnets, so keep your eye out for one when the motor wears out].
The "Smart Drive" motor consists of a driven outer cover with the magnets embedded in the plastic, and an inner stator made up of copper wire wound metal cores. Because the inner stator is stationary there is no need for brushes - another big plus. To save a lot of work adding bearings, it is best if you can at least get the tub part complete. The whole lot is better as you can make the vane out of the outer case (by folding around the edges to stiffen things). To hold the bearings you need some of the plastic around the shaft. Be generous to start with and bore holes well out, and then cut out with a jig saw. You then can bolt this lump of plastic - after you have trimmed it - to a suitable piece of metal or wood that in turn can be bolted in the correct position to take a belt from the brake drum.
Remember, all the plastic is NOT UV proof, so everything has to be painted and covered.
Unfortunately the motor has to be modified before it becomes useful to charge batteries. If you can cut and solder wires, you can make the modifications. If you can't cut and solder wires, then may I suggest that it is worth learning, or worth getting someone to help, as the output, stability and reliability of a permanent magnet generator far exceeds the previous two options. I have seen photo's of where people have used "screws in the end" electrical connectors instead of solder, so that definitely an option if you can not solder. If you turn the spindle in its original condition, that is, move the magnets over the coils, you get about 90 volts AC (could be lethal) at less than an amp. Almost useless for charging batteries. However, with modifications, the results are far different. There are diagrams available on the Internet that go into great detail how to make the modifications, and that would be a good place to start (*www.thebackshed.com - click on Homegrown power, scroll down to ">>View more projects", then click "The F&P Smart Drive"). However, it may be possible that someone is reading this in printed form, and doesn't have Internet access. I will try to put into words what the diagrams are saying. Also see diagram below.
If you undo the centre plastic nut (turn anticlockwise) and pull the magnets off from the stator, you can easily see that there are 14 magnets and 42 coils. This means that coil 1 and coil 4 and coil 7, and so on, are under the same polarity of magnet. To put it another way, every No. 1, and every No. 2, and every No. 3, in each set of three coils are behaving in the same way, but in one set of three, 1,2 &3 coils are in a different phase. It just so happens - fortunately - that two coils in series will produce enough voltage to charge 12, 24, or even 32 volts in a pinch. Obviously you get more amps at the lower voltage, but the watts are about the same. It is a very useful battery charging source, and will give 20 amps at 24 volts if the fan and gearing are reasonable.
You may be thinking that all that needs to be done is to put each No. 1 and No. 2 and No.3 of two groups of coils in series, and then each group of two in parallel, and the job is done - and that is exactly right. But doing just that is not easy - especially the first time. However, in English it goes something like this.
Because we have three coils under each magnet at the same time, each coil in a set of three coils are at a different phase at any point in time. This is why the first coil in a set of three has to be in parallel with the first coil in another set of three. They are both "pushing " or "pulling" at the same time. And the second has to be connected to the following second - and so on. Then to keep the phases in their correct group, the output from the first coils in the first set and second set of three, can only be put in parallel with the output from the first coils in the third and fourth set of coils. If we just look at the first six coils after the wires are cut to isolate them, we have six ends. Two from a pair of No. 1 coils, two from a pair of No.2 coils and a pair from the No.3 coils. Each of these "ends" have to be soldered onto a separate "bus" or copper wire, and thus we end up with six wires running around the base of the coils.
If each pair is connected to, say, a 35 amp bridge rectifier (available from Dick Smith) and the three positives and three negatives of the rectifier are wired in parallel, then the unit will produce the maximum power at maximum speed. However, the wind does not, usually, blow at maximum speed all the time, so the fall-back option is usually better. Three of the "ends" are joined together, and the other ends go to just one terminal of a bridge rectifier. This spreads the load/heat evenly, but two could be used with one terminal spare. This is called a Star configuration. The positives and negatives are again wired in parallel. This gives a better result at lower speeds. Some people have got fancy, and have come up with a way of switching between the two, depending on the wind speed, but I haven't tried it yet. The bridge rectifiers (or what ever rectifiers you use) should be mounted on metal or aluminum to act as a heat sink. Remember, on a hot summer's day, things are going to be pretty warm up there.
A word about the cutting and soldering.
The six wires forming the "bus" at the base of the coils need to be heavier than the coil wires, because it is carrying more current. Also it needs to be insulated so that the bus wires don't short. This means that you are going to have to cut the insulation off (I use a Stanley knife) at the right place to solder on the leading and trailing wires of the two coils. You could use plain or stripped copper wire, and then slide on some heat-shrink after you solder each joint. The coil wires are resin insulated, so you are going to have to scrape this off before you can solder. I use a Butane flame on the cut end (softens things up a bit) and then squeeze and pull with emery paper, changing the position each time. Try to wrap the end of the coil wires around the bus wires before you solder them to give some extra mechanical strength.
It ends up not looking pretty, but I squeeze silicone between the coils (especially where the coil wire has been cut and might be a bit loose) and over the bus wires to make sure nothing moves. Because it is generating AC, any movement will eventually cause a loose wire to break under the constant flexing as the direction of current changes. Keep the wiring out of the way of the rotating shaft and magnets.
When you have finished, put the magnets over the stator, and get someone to spin it, or use an electric motor or drill, and check that the AC voltage out of the three pairs is roughly the same, and you are getting DC out after the diodes. With hand spinning it will only be a few volts, but that is enough to show that things are working.
The above method of modifying the Fisher and Paykel washing machine motor is the best for when you are driving the generator off a brake drum with at least a 2:1 gearing up.
However, if you are just using a straight through drive - no gearing up - then either of two other modifications are better. The method of cutting the wires between the coils and soldering them onto a bus is the same, but the configuration is different.
You still have to deal with all No.1 position coils in the groups of three, but instead of having two in series, and then all those twos in parallel, you have two groups of seven, and then have them in parallel. Obviously this involves a lot less cutting and soldering, so that can be an advantage if you are just starting out. This configuration has a higher back EMF (resistance to turning as the voltage goes up) so is very good for remote locations or weekenders, where you need the generator to take care of itself. To put it another way, because the resistance to turning increases with speed, it is sort-of self governing. Sure, it does turn faster in a very strong wind - and puts out more power - but it won't go into free-wheel and self destruct. It will charge 12 or 24 volts. It is also the best configuration for light winds, and will put out 1 or 2 amps in a very light breeze - and long before anything else is even turning. On the negative side, it rarely puts out more than 7 amps.
A bit more complicated.
This one looks odd at first, but it is the one I prefer for 1:1 gearing and wanting 24 volts. Everything is the same as for the first one, but this time the configuration is 5 + 5 + 4 (it must add up to 14 - the number of magnets). Looking at the coils in groups of three, you have 5 No.1's in series, going into the bus, with 5 more, and then 4, and the same with the 2's and the 3's. Since you have three coils in each group, and to spread the heat load, you could have 4 + 5 + 5, 5 + 4 + 5, 5 + 5 + 4. Sure, the section with four does not pull its weight until it reaches the same voltage as the fives, but once it does, all the individual coils are working about the same. The unit I have puts out 10 amps at 24 volts in a strong wind, and around 14 amps at 12 volts.
The reason you have to change the configuration between gearing up and 1:1 is that two coils do not generate enough voltage at lower speeds. Seven creates the most voltage with some reduction in power, and the 5+5+4 is a good compromise. I have not tried it but 7+4+3 could work well in a 1:1 situation if you use shorter blades - to increase the speed - and more blades - to increase the "push".
If you want maximum power from your F & P, put two coils in series, and all the two's in parallel, and use some sort of gearing up.
This can be as easy or as hard as you like to make it. The advantage of doing it the hard way is that you have almost zero maintenance. For the easy way you have to unplug the wires coming out of the generator and untwist them every few months or so.
The commercial type of hub that allows the current to be transfered via carbon brushes - and thus the fan can point in any direction - while not impossible, are difficult to make. If you want to have a go, then get a good look at an old Dunlte or other commercial type. For now, I will ignore that type as it is usually beyond the resources of a DIY generator builder. That said, if the wind destroys a generator, usually the mount survives, so you might be able to pick one up second hand - as I did.
For the home-made mount, you need about three feet (1m) of solid metal pipe (pipes come in many grades, steam the best) that just fits inside another piece of pipe about six or eight feet (2 or 3m) long. To hold up the inner pipe you need to weld a collar around the pipe just under the lowest point where you have attached the generator. Then weld a collar around the larger pipe at the top. To reduce friction you apply plenty of grease. Of course, if you can afford it and can do it, putting an inner cone wheel bearing out of a truck between the two pipes will do a much smoother job. Putting something "softer" like nylon or brass between the collars may help.
The next bit I feel is important, but most generators do not use the principle. The axis of the axle needs to be positioned 2 to 4 inches (50 to 100mm) to the right or left of the axis of the pipe. The reason I do this is so that the fan can twist out of the wind in heavy gusts, and this takes a lot of pressure off everything. The amount of "twisting" is determined by the length of the vane and the size, and this can only be determined by trial and error. I would start with a four to five foot (1.2 to 1.5m) vane pipe or square section, and a 1' by 2' (0.3 by 0.6m) vane. If it twists too much, increase the size of the vane. If you want more twist then reduce the size.
The wires then are run down the centre of the pipe. Near the ground or where it is easy to get at, you need an easy to undo connection . If you have nothing better, you van use two extension cord plugs, and ignore the earth pin. Because the winds go round and not back and forth, the wires will get twisted over time, so you will need to unplug and untwist the wire every so often.
The Gin Pole is needed to give the chain more angle so it can lift rather than just pull.
of wiring changes required for the stator coils.
Star gives more voltage at lower speeds.
Delta gives more power at higher speeds.
If you are just starting out, I would go with the star configuration. This gives good results for a range of winds.