wind generator
zapmaster
Posts: 54
Im looking for some insight. Im making a controller for my wind generator.
This is my plan. Use a small s12 car alternator. Have the stator rewound and connected in a Y connection. Use the field voltage to control TSR(Tip Speed Ratio). The controller is going to be a parallax propeller. The inputs to the propeller are anemometer and tachometer. The output of the propeller is going to be a variable dc voltage regulator controlling the field. I will be using the rectifier with the generator. Tests so far have been encouraging. I have a test bench with a dc motor on a variable supply. I can control speed of the dc motor to simulate the wind. The field will consume 0-43 watts. The hope is to have a very efficient 500 watt generator. The output will be 12 volts. Any thoughts?
This is my plan. Use a small s12 car alternator. Have the stator rewound and connected in a Y connection. Use the field voltage to control TSR(Tip Speed Ratio). The controller is going to be a parallax propeller. The inputs to the propeller are anemometer and tachometer. The output of the propeller is going to be a variable dc voltage regulator controlling the field. I will be using the rectifier with the generator. Tests so far have been encouraging. I have a test bench with a dc motor on a variable supply. I can control speed of the dc motor to simulate the wind. The field will consume 0-43 watts. The hope is to have a very efficient 500 watt generator. The output will be 12 volts. Any thoughts?
Comments
Why would you want to re-wind a standard product ?
Switch mode load matching would surely be easier, especially with even moderate processing power.
Here is a good example of choices
http://cds.linear.com/docs/Solutions%20Manual/High_Performance_DC-DC_Controllers.pdf
Search that pdf for 'Equal' to see designs that allow VIN Above, Below or Equal to VOUT
What I have learned so far. Car alternator need to turn very fast. Wind turbines need to turn slow. Stock alternator are wired delta wiring them in a Y will up the voltage by 1.732 but decrease the current by the same factor. That was not enough volts per rpm. The only way to boost the voltage is to have more windings. What I found is a 3 blade generator has a tsr of 5. So if the wind is blowing at 12 miles a hour the blade tips can only go 60 miles per hour. 60 inch blades that is 67 rpm. This prevents the turbulence from the blade in front from affecting the efficiency. That is why the big turbines go very slow. 6 rpm. you can get them on ebay. got this one for 33$.
thank you if i can not get the voltage / rpm up to 12 vdc+ i will keep this i mind.
what is to benefit? loose efficiency in the restance of the windings or in the dc to dc converter?
The benefit of a design that allows VIN Above, Below or Equal to VOUT, is you no longer have a sudden cut-off at lower wind speeds.
That gives you a significant 'area under the curve' gain in light/gusty lifetime use. - plus you need regulation anyway, and doing that only via field is less optimal.
With the ability to vary both Field and load points, there is probably a MPPT operating point you can seek.
Most of those use a simple 'vary and check' algorithm, and a Prop would handle that very well.
I am also working on a wind generator controller using the propeller. I hope you have realized that there is a *lot* of misinformation about wind generators on the internet. I recommend that you go (as fast as possible) to otherpower.com and scoraigwind.net to get more information about DIY wind generators in general.
You are right that wind generators turn too slow for a standard car alternator, and that if you want to use them, you have to rewind them. However, in practice it is very hard to get 500W out of a converted car alternator. Most DIYers use a permanent magnet generator that they build themselves; it will be much more robust and reliable, and easier to tune it to what you need. The only problem is that neodymium magnets have gotten very expensive over the last couple of years, and you'll probably have to go with ferrites, and most of the info on the internet is about neos.
While a TSR of 5 is a good starting point, most find (particularly using ferrites) that spinning a bit faster is more efficient for the alternator, and a TSR of 7 or 8 makes things easier.
As far as MPPT goes, you probably do not want to attempt this. Realize that even utility-size wind generators (with their attendant highly-paid engineers) do not use MPPT. The problem is that each time you increase the duty cycle of your controller, you will always get a temporary increase in power due to prop inertia, so if you let your controller seek the MPP using perturb-and-observe, you will get a runaway feed-forward loop that will either eventually stop your prop, or blow something up, or both.
What you can do is something that I call MPPF, or Maximum Power Point Following. It turns out that you can calculate the best duty cycle for your RPM, Battery Voltage, and desired TSR, from knowing the particulars of your wind generator, including V/rpm, generator impedance, and line impedance. There is no power feedback for this, since you are not trying to maximize power (other than wishing the wind would blow harder !); the effect will be the same as if you were, but the broad TSR plateau allows you to estimate the power pretty well with calculations, and you will be close enough to the MPP that it doesn't matter. When you are ready for this, I can show you the calculation.
However, particularly for such low power levels, implementing a DC-DC converter (specifically, a buck converter) will probably not be worth it unless you just want to, which is certainly a fine enough reason to pursue this ! Most just connect the wind generator to the battery, and then use a diversion controller (which could certainly be a DIY project for a propeller, but is available commercially for about $150) to divert extra power once the battery gets full. You will not have a constant TSR with this method, but it is simple and safe.
I'm not really an expert on this (not many are, since all the smart guys realize there's not much money in hobby wind generators), but I'm a mechanical engineer with a strong electrical bent that understands wind generators pretty well. Feel free to contact me for more information, or go to the forums at otherpower.com (fieldlines.com). There's a lot to learn about wind generators before you get to the part of the controller where you would ask questions on this forum.
Good Luck with your project,
David Voss
I think the primary reason that the big turbines turn relatively slowly is to keep the tip speed at practical velocities (around 120 mph). That's closer to 20 rpm for a turbine with a 50 meter rotor diameter.
The largest wind turbine, the Enercon E-126 has a rotor diameter of over 400ft and can turn at over 11 rpm. That's a tip speed of of over 160 mph. I would expect that they would reduce the max tip speed if used in areas where noise is a problem.
But in the last ten years what I have done is a lot of research, and perhaps this is why I haven't built one yet. I have come to the slightly cynical conclusion that the alternative energy industry is built on lies. Solar, for instance, love to give you their wattage at midday on a sunny day. They don't give you the output on a cloudy day, or at midnight. So the real wattage averaged over 24h is about 20% of what is on the product.
Wind does blow 24h a day most day, but there are other mistruths in the wind energy sector. I managed once to get the real data from a water pumping windmill (which has a pretty good wind to mechanical efficiency) and got a ballpark figure of 30 watts per square metre. That is a radius of around 56cm which does give you the right sort of tip speeds for a car alternator. However
You will be lucky to get enough power to run the field. Make it bigger and it runs slower.
Another thing the engineers tell us is that for wind loading, one square metre is one ton of wind loading for a typical gale that might come along every few years. It is the cube law thing with wind, and you have to design a turbine to extract power from slow wind speeds and at the same time survive a gale. So in some ways, designing a wind turbine is not about maximum power point tracking, but more about feathering and presenting as little area to the wind in a gale. It is more about throwing power away than extracting power.
Having said all that, I thing there is great potential for smart controllers like the propeller. Think of an old sailing ship, with all the crew putting up and taking down sails and varying the power extraction. Take down all the sails to survive a storm.
I've got a shed full of alternators and PM DC motors and other mechanical things. One day I am going to build something. I think it needs to be big (to extract a decent amount of power, and that means a radius of many metres), it will run slow so needs gearing up, it needs to be light, it needs to be easily servicable, and it needs some smart way of feathering or decreasing surface area as needed when the wind speed is too great. Oh, and I have no desire to build a standard multiblade water pumping windmill as I once had a long conversation with a guy who services these and he said they look much bigger when you are up on the deck and there are sharp 12ft blades whizzing past just near your head!
I've got some crazy homebrew ideas but like most practical things, the proof is in the building and how well they actually work in practice. But in general, small turbines don't pay back the cost of their materials within a reasonable time frame, big turbines need tractors etc to hoist and need a gearbox at the top of the pole, so my ideas have turned more to vertical axis turbines as all the gearing can be at ground level. I kind of like Darrieus turbines but I think the blades need to be stiffer to avoid vibrations at certain speeds.
I'll second the comments from David Voss. There is a lot of practical advice at the otherpower website. Homebrew windmills using rare earth magnets are popular and the impedance matching is probably better than for a rewound alternator.
Many years ago I studied this fairly in depth, but since have forgot most of it. However this I do remember, if you are seeking information on this subject, this link is probably the best place to start.
http://otherpower.com/otherpower_wind.html
Bruce
EDIT: I now see that someone beat me to it
http://www.otherpower.com/images/scimages/1431/Alternator_Secrets.pdf
Additionally, the attached partial favorites may help you find some helpful information. You should be able to import these into your favorites if you are using a WINDOWS OS
Bruce
There is a general trend away from gearboxes, and with MOSFETS getting ever cheaper, you can move the load operating point, to almost anywhere you want, electronically.
MOSFETS + Prop, lets you roll almost any switching topology, even synch rectification.
Gale handling remains a significant issue, and I like the pivoting blade designs, which have simple inbuilt feathering.
There is even a single blade wind turbine, if you want your design to turn heads....
https://encrypted-tbn0.google.com/images?q=tbn:ANd9GcRrKdbXix9KEvMrI8V_dBPRIkBNIdS1x_FkV3O6dyg01KnvYyd8ng
This one has the pivoting blade.
What you can do is something that I call MPPF, or Maximum Power Point Following. It turns out that you can calculate the best duty cycle for your RPM, Battery Voltage, and desired TSR, from knowing the particulars of your wind generator, including V/rpm, generator impedance, and line impedance. There is no power feedback for this, since you are not trying to maximize power (other than wishing the w
David Voss[/QUOTE]
Thank you
Why are permanent magnet better no generator AC or DC where I work use them we use wound field coils.
I plan to use the field for mppf. I will have to do some research on mppf. They are reliable on our cars.
Increase the field current will cause the voltage to increase but the battery will lock it into 12v so the current will increase causing the blades to slow. The higher the current in the field the more lines of flux the more lines crossing more windings in the coil the more voltage.
Things are working good on the bench. Need to hook up the propeller next, parts are coming in the mail.
I want to output the data from the generator using a xbee to my webpage.
This project is just for fun right now most of my parts are free. I get the towers for free. And have the tractor to stand them up an down. The generator I have now produces 70 watts now with no control that will run the feild . So the blades have the ability. 200 rpm at 4v field = 15 watts.
Thanks for the info.
That should work, but maybe a backup as well, like feathering at 90 degrees or tilt up?
PM vs field windings. An interesting discussion. I guess it is the power you lose in the windings that you can't get back. But the ability to alter the field and hence the load has some advantages - eg you can avoid cogging at low wind speeds, and you have more control over extracting the energy, plus tesla for tesla, I think copper wound round iron is cheaper than rare earth magnets http://forums.parallax.com/showthread.php?142470-Got-magnets-1200-lbs-of-force
Stall ? This could be a challenge, as you would also need to 'short' or heavily load the windings, and the blades will still spin, but you hope at a safe speed, and with an i2r loss also within spec.
Some useful power may be possible in these conditions, as a bonus.
Worth a test.
There are still large lateral forces, & likely resonances, even if the RPM has not gone off the scale.
"Stall" means that you have a large enough load on the prop that the actual RPM is far lower than the design TSR, so the angle of attack on the blades is far enough from optimal that any additional wind force can be counter-acted by the electrical load and will no longer cause an increase in RPM. In practice, you need a *very* efficient alternator to do this, because the current is far above the design value in order to induce stall. That is one reason that you may want to match the load with a DC-DC converter, rather than a wound field. There are two other ways of inducing stall that are much more common than overloading your coils. In large wind generators, they alter the angle of attack by pitching the blades; Jacobs is a small wind generator that uses a mechanical flyball governor to do the same thing. Another way is furling, which is angling the plane of the blades out of the wind; almost all small wind generators do it this way, and you would be well advised to use this as a backup even if you plan to do it with prop loading.
I'm not saying you *can't* induce stall with your alternator; it's just that, in order to do so, you will have to over-design it to the point that you could have had 2-3 times larger blades (and much more power as a result) with the same alternator. It's just not the most efficient way.
If you want to go the route of modifying the field, I would recommend using a real 3-phase induction motor designed (probably rewound, also) for low speed and a higher current. You can then use a microcontroller such as the propeller to drive a 3-phase synchronous rectifier, and adjust the field by varying the electrical RPM relative to the prop RPM. Note that you can also do this with a permanent magnet alternator.
There is an article on otherpower.com about furling (and many of the step-by-step instructions for building a wind generator discuss it), and you should search the forums on fieldlines.com for the story "Matching the Load" by the user "flux", who is a storehouse of information. This is my last post on this forum about this; while I could talk about it all day (and sometimes do !), it's really off-topic here. There are many more knowledgeable people about wind generators on the actual wind generator forums, so you are really just spinning your wheels on a microcontroller forum.
As for cogging, PM generators definitely cog; I've felt it myself. They just don't cog enough to matter at the load they're able to draw on the turbine.
Thank you this is somthing i will have to test. i could over heat the stator before i stall the blades or overcurrent when i do. my back up to slowing down a runaway was going to be a centrifugal clutch. I will look into the furling.
Could not have said it better. It is a very romantic notion to harvest "free" electricity, but the stark reality is that wind power ends up costing approx. 2x and solar roughly 10x conventional methods.
That said alternative power generation has its place e.g. too far from power lines etc. It will probably be decades for all of the factors to converge to be feasible. I do find solar/wind to be very interesting from a hobbyist perspective.
I am following this thread with great interest, good luck zap master!
A prop chip could manage the switching needed for power conversion, and the loading needed to stall - worst case is a simple short, but you might prefer to heavily load the alternator, (far past optimal) but extract just enough energy to power the field. (via step-up).
You do not really want to be draining system (battery) energy, to keep control - control should work without a battery connected.
That, combined with a simple spring-pivot system for furling may be enough.
You could test this to storm levels, with a Generator mounted on a truck. Hurricane levels you might want to lower the whole unit, as even a furled model, can still get wiped out with flying debris !
A friend of mine, a retired Boeing engineer, built a solar panel installation on a trailer that he towed behind his motorhome. I don't recall the exact figures but he spent around $4k on it and after a year he calculated that he produced less than $20 of electricity. He admits that it was not the most efficient system but even at ten or twenty times the output it will take a long time to break even.
Electrically, this has merit, and I've seen Tidal Generators done using outer-ring induction.
For an example :
http://www.rechargenews.com/multimedia/archive/00032/Tidal-power-Nova-Sco_32957b.jpg
However, for a wind generator, where all this has to go up on a pole, and add wind loading, and survive gusts, it is a mechanical recipe for disaster.
The problem is stiffness - both of the blade and of the surrounding structure. The air gap might only be a millimetre.
But in a way, the axial flux motor is a variation on this theme. The 'standard' DC motor has a magnet and the rotating wires sit inside the magnet. It rotates very fast with low torque.
An axial flux "pancake" alternator moves things out further so you get higher torque but slower speeds http://www.gotwind.org/diy/Axial_Flux_Alternator.htm
That link has a nice photo showing the air gap. The thick iron backing is both for stiffness and to complete the magnetic circuit.
This is another example http://warlock.com.au/10kwgenerator-abstract.htm and two things in that article that stood out for me are the costs of the magnets ($500), and the 640kg force between them as you put it together.
From a magnetic point of view, ferrite or cold rolled silicon steel laminations (ie from a transformer) would be better than solid steel as it would reduce eddy current losses.
The homebrew turbines seem to use high strength magnets and often air cored coils. An interesting article here http://www.usmagneticmaterials.com/documents/HarvestingWind.pdf talking about the economics of rare earth magnets vs copper and steel DC powered electromagnets for the windings, which goes back to the original post talking about a rewound alternator.
It is interesting to ponder the costs of all these materials, particularly in terms of "payback time".
this is a very interesting thread. I wonder, if a bicycle motor could be used as a suitable generator, because it works with relatively low speed and might be affordable.
Example: http://www.bike-emotion.org/Bafang%20SWXH%20250W%20Hinterradmotor.pdf
If you look for a tubine, that is easy to build and starts at low wind speeds, a Savonius Rotor might be worth consideration.
Christof
I accept that this design will be challenging with regards to stiffness and clearances. Have to think more about that.
Erlend
Certainly should have economy of scale on it side
It also matches 36V (motor) better which is more cable and MOSFET friendly than 12V.
Biggest challenge is the wire exits the axle, which needs two supports, and it expects the OUTER body to spin ?
Connections are the 3 phases, plus the Hall sensor, and brushless, so is more suited to some smart commutation.