12V Wind Turbine - Voltage question

[OmCheeto]

Oh dear. That video is from 2006.
Vaporware!
Glad I never tried to build one.

It did strike me as a bit of an engineering nightmare.
But beautiful to watch, in a physicsy kind of way.

 

[Averagesupernova]

I have rebuilt a few older delco alternators and what I find is this: There is a 6-diode bridge off of the stator that is the main rectifier. This feeds the voltage regulator and the main battery post. The output of the rectifier assembly is the point that the regulator looks at to determine whether or not to increase the field current in order to raise the output voltage based on a comparison to an internal voltage reference. Then there is a diode trio that connects from the stator to another terminal on the regulator. This I consider a power terminal. If this terminal has no voltage on it the regulator will not allow field current in the rotor. This terminal ALSO is fed through the dash board indicator when the ignition is turned on. So what happens is when the ignition key is turned on the regulator is powered up through the light bulb. This is enough to tell the regulator to turn the field on based on the battery voltage. When the alternator comes up to speed the diode trio rectifies the output from the stator and the regulator is now being powered by the diode trio. The dash indicator will no longer light since it has about 12 volts on each side of it. At this point the bulb can be removed with no change in operation of the alternator since the regulator is being powered through the diode trio.

 

[jim hardy]

If this terminal has no voltage on it the regulator will not allow field current in the rotor. This terminal ALSO is fed through the dash board indicator when the ignition is turned on. So what happens is when the ignition key is turned on the regulator is powered up through the light bulb.

Good description.

I'm accustomed to a resistor in parallel with the lamp. I both gives you alternator should lamp burn out and it let's "bootstrapping" start at lower RPM than with lamp alone.

 

[owen_a]

@OmCheeto thanks!

I was actually going to build a VAWT using one of these alternators (I think they're delco, I'll check later). I used a bicycle wheel and a 200cm fan belt to get my ratio. I think it was 1:10. The only thing that stopped me, is the fact that it wasn't easy turning that wheel. You needed a little bit of force which I doubt the wind could do unless if it was really windy. This is when I thought that a HAWT might be a lot better since the force against the turbine is stronger. However, the blades would need to be quite big - you did the math above.

What if you had a resistor in series with the field coil, so it reduced the amount of current in order to self excite? Would it still self excite but reduce that initial torque? There was a site which I was reading a few months ago, and they used an alternator as a generator for a water wheel, and used a resistor to do something like this.

 

[OmCheeto]

@OmCheeto thanks!

I was actually going to build a VAWT using one of these alternators (I think they're delco, I'll check later). I used a bicycle wheel and a 200cm fan belt to get my ratio. I think it was 1:10. The only thing that stopped me, is the fact that it wasn't easy turning that wheel. You needed a little bit of force which I doubt the wind could do unless if it was really windy. This is when I thought that a HAWT might be a lot better since the force against the turbine is stronger. However, the blades would need to be quite big - you did the math above.

Actually, I didn't do any math. I just plugged numbers into the "computer" until I got my answer.
The equations are on another page:

Power generated from the wind
The theoretically available power in the wind can be expressed as
P = 1/2 ξ ρ A v³
where
P = power (W)
ξ = efficiency of the windmill (in general less than 0.4, or 40%)
ρ = density of air (kg/m³)
A = area wind passing through perpendicular to the wind (m²)
v = wind velocity (m/s)

If you look at the variables in the equation, there's really only one that you have control over, and that's the "area" of the wind turbine.

 

[Merlin3189]

"If you look at the variables in the equation, there's really only one that you have control over, and that's the "area" of the wind turbine." I think this a very important point. There is a danger of thinking, 'if only we used a different alternator, or geared it up, or down, or whatever', then we could make it work. But the essential point is, there is only a certain amount of energy/power available from a moving mass of air (and apparently*, we can only extract about 60% of that.) Nothing you can do with your alternator can alter this.

All you can do is to optimise the extraction / conversion process, so I think you need to look at the steps in the process and see which ones can be improved. Two items related to the alternator which I think could be changed to help you are both electronic.

What might help you most with your main problem (getting output at low wind speed) is synchronous rectification. Instead of using diodes, as shown in the circuit of #33 and used universally in automotive systems, you can use mosfet transistors (and associated control circuitry.) Diodes have a roughly constant voltage drop of 0.7V, so that two in series in the standard circuit drop 1.4V. To get 12V output you therefore need to generate at 13.4V and waste 10% of your power in the diodes. Mosfets do not have a constant voltage drop, but a very low resistance of only about 0.01 V per amp. So at low currents, say up to 10 A, the voltage drop is less than 0.2 V, wasting only 1 - 2% of your power. At high currents you eventually lose out to diodes, but then you are not so worried about losing a bit.

Another thing you can do with controlled transistor rectification, is simply to switch it off at low speeds. This could solve your starting torque problem: if no power is being extracted, there is no electromagnetic drag and your alternator should rotate freely, limited only by friction.

I think the reason this does not seem to feature in many amateur wind turbines, is that automotive alternators usually have diodes built in and possibly regulation as well. But I guess it would not be too difficult to remove or bypass them.

If you get away from automotive alternators and have them made specifically for your purpose (possibly, increasing the cost 10x of course!) then you can also do things such as SophieC's suggestion of increasing the poles to make it function at lower speed (but higher torque.) Doing so does not give any more power, but by avoiding gears / pulleys will reduce frictional loss.* "Betz Law
In 1919, a German physicist Albert Betz, based on conservation of momentum and energy, he proved that the maximum possible energy that can be derived from a wind turbine cannot be more than 59.3 percent, or 16/27 of the potential energy in the wind.
In practice, no wind turbine has ever achieved the Betz limit."

 

[owen_a]

What might help you most with your main problem (getting output at low wind speed) is synchronous rectification. Instead of using diodes, as shown in the circuit of #33 and used universally in automotive systems, you can use mosfet transistors (and associated control circuitry.)

Using MOSFETS? hmmm, I've never used anything other than diodes for rectification. Something like this maybe? (second picture)

 

[jim hardy]

Something like this maybe? (second picture)

Clever, that self biasing... No timing circuity required to synchronize them.

 

[Rive]

I have currently got my turbine connected to this rectifier (although proven to be inefficient to it's 2v+ drop on the diodes), and then connected to a 'boost buck' which 'steps up' the voltage to 14.4v which I variably set. I then had the output connected to my solar charge controller so I can see if it's outputting any power when it's spinning, and it reached 14.4v @ 1A - though there was barely any wind at that moment, so it didn't rotate as fast as it did before.

If you change the rectifier bridge to some schottky type you can save some on the voltage loss (without spending too much on special rectifiers), but I think the root of the problem lies elsewhere. You already nailed it with using a charge controller, but on the other side you also lost it with using the step up converter.
A charge controller (at least a good one) will try to optimize the power pulled based on the input voltage: higher voltage comes in -> it'll draw higher current (it 'thinks' that when there is more light/wind/whatever there will be higher voltage, so it'll pull more juice from the source).
But with the step-up you already fixed the voltage -> it'll always work with the current which belongs to the 14.4V -> the system will deliver only the fraction of it's maximal potential.

Also, as this setup will draw only a ~ fixed power, it cannot effectively slow down the rotor in high wind: it'll always overspeed (this is what you are experiencing, right?).

I think what you need is simply to match the output voltage range of the turbine and the charge controller - and forget the step-up.

 

[jim hardy]

What if one collected his power at 6 volts instead of 12?
Seems to me he'd halve his speed requirement
and a slow turning propeller has better efficiency , as the Wright Brothers figured out.
http://wrightstories.com/propeller-design-demonstrates-the-genius-of-the-wright-brothers/

 

[OmCheeto]

If you get away from automotive alternators and have them made specifically for your purpose (possibly, increasing the cost 10x of course!)

As a perpetual pauper, that's never been a viable option for me.

then you can also do things such as SophieC's suggestion of increasing the poles to make it function at lower speed (but higher torque.) Doing so does not give any more power, but by avoiding gears / pulleys will reduce frictional loss.

Hence, why I hobble together old junk, and do the sciency stuff:

The last experiment was to determine how much power was being lost by the fan belt.
It turned out to be around 100 watts.

I decided that either direct drive, or a more flexible serpentine belt would be more efficient.

[ref]

 

[jim hardy]

Hence, why I hobble together old junk, and do the sciency stuff:

that's how real progress happens.

I kept this poster over my desk for years.

picture courtesy of http://www.greenwichworkshop.com/details/default.asp?p=2106&t=0

 

[Howard]

Sounds like you need to gear up the drive between propellor & alternator. See the alternator performance curves to get the correct speed. I designed & built one of these a while ago that still works well. I'll dig through my piling cabinet; keep your fingers crossed.
Regards, Howard.

 

[Merlin3189]

Owen & Jim point to a mosfet circuit which I had not realized could be done and seems a good simple idea as an immediate replacement for a diode bridge.
I had been think of a more complex circuit like this or an even more complex one (which I can't find a link for at the moment) which included input from a regulator. I must say though that I have not built nor used any of these circuits, I am simply aware of the technique and thought it might be another step improvement to the system.

I notice a comment, " it cannot effectively slow down the rotor in high wind: it'll always overspeed (this is what you are experiencing, right?)." and wonder what the issue is here? It seems to me that, if you draw current to slow the turbine by loading it, you risk overheating the windings. Isn't that the problem of high speeds? Would it not be better to limit the output current at high speeds and let it run as fast as it wants? If mechanical damage due to centrifugal effects is the worry, would it be better to have some sort of mechanical governor to feather the blades say? (Maybe mount the blades on a spring-loaded screw, so that as centrifugal forces pull them outwards - physicists please excuse my language - they twist slightly.)

Oh, & BTW OmCheeto, "As a perpetual pauper, that's never been a viable option for me." i heartily concur ! I wasn't so much suggesting you did this, as pointing out that difficulty if you wanted to have a tailor made alternator. All respect and encouragement to those who are attempting to do this themselves.

 

[jim hardy]

If one accepts the complications of a boost/buck regulator he can collect several times the alternator's rated power by letting it run at higher speed,
Higher voltage at rated amps = more watts;

and he can generate in low wind at lower RPM. where he can only make less than 12 volts .

Owen's self biasing mosfet rectifier circuit should enhance efficiency at low voltage

that MIT link in post 24 should really enhance high wind performance.I predict some young folks will be making smart rectifiers with Arduinos soon enough.

A 150 amp car alternator at 40 volts would absorb 6kw divided by its efficiency,
surely that'd keep most homebuilts down to safe speed short of a hurricane?

 

[sophiecentaur]

Imo, it would be the turbine design that would impose the final limit on the overall power obtained. That's a far more difficult problem to solve than making a good rectifying circuit.

 

[OmCheeto]

Oh, & BTW OmCheeto, "As a perpetual pauper, that's never been a viable option for me." i heartily concur ! I wasn't so much suggesting you did this, as pointing out that difficulty if you wanted to have a tailor made alternator. All respect and encouragement to those who are attempting to do this themselves.

A windmill is now #743, aka "at the bottom", on my list of random projects to build.

Imo, it would be the turbine design that would impose the final limit on the overall power obtained. That's a far more difficult problem to solve than making a good rectifying circuit.

I ran across a 59 page paper yesterday on wind turbine tower design. Pages 23 through 29 list just the variables involved.
6 pages of variables.
I can't even solve a problem with 6 variables, period.

But I do like Merlin3189's idea of blade feathering.

 

[Rive]

I notice a comment, " it cannot effectively slow down the rotor in high wind: it'll always overspeed (this is what you are experiencing, right?)." and wonder what the issue is here?

In high wind you not just 'can' draw the rated power of the turbine, but you also 'have to' draw that power to keep the RPM within the specified limits.
Your actual setup cannot draw that power, so the RPM goes out of the specification.

 

[Merlin3189]

Imo, it would be the turbine design that would impose the final limit on the overall power obtained. That's a far more difficult problem to solve than making a good rectifying circuit.

Absolutely agree. Pretty much everything about the electrical conversion process is understood. How we apply it depends mainly on costs. But the wind! I have read that Einstein hoped that when he died, God would explain turbulence to him. When I read about something like mosfet rectification I can see at once how it could help, but if I see a picture of a wind turbine, I often have no idea whether it's better or worse than another, or even whether it would work at all! But being so complex does make it a plausible area for experimentation.

you also 'have to' draw that power to keep the RPM within the specified limits.

I think maybe I missed something in the posts along the way, but I'm still not sure what this RPM limit is about. I'm not saying there isn't good reason to limit the rotary speed, just wondering what it is.
If you do draw power to keep the RPM down, you are still putting stress on the mast / mounting (which may or may not be your critical factor.) If you let the rotor freewheel, then you do reduce this stress, though your centrigfugal stress rises. Depends what your main concern is.

 

[sophiecentaur]

I'm not saying there isn't good reason to limit the rotary speed, just wondering what it is.

Is it not to avoid it tearing itself off its mounts and chopping up everything it meets with? I must say, feathering is a lot safer feeling than electric braking but it's a much more complicated mechanism. I guess a cheaper half way house could be to have a method of steering it off the wind with the vane when the speed gets excessive - an electromagnetic clutch could do the job but you'd need to have a pretty stable control system or it could end up spinning round and round on its vertical axis.

 

[owen_a]

I guess a cheaper half way house could be to have a method of steering it off the wind with the vane when the speed gets excessive

My turbine actually does this. If the wind gets to high, it steers it out of the wind, and if the blades rotate to quickly, the electromagnetic brakes kick in.

 

[sophiecentaur]

My turbine actually does this. If the wind gets to high, it steers it out of the wind, and if the blades rotate to quickly, the electromagnetic brakes kick in.

IS the steering done electrically or with some mechanical system? Sounds good.

 

[owen_a]

It's done mechanically, I'm presuming by the design of the Wind Vane since it has a gap in the centre of it. Probably aerodynamically built that way.

 

[Rive]

Depends what your main concern is.

Max out the power available from the given turbine?
Without appropriate load the turbine is overspeeding. With a well matched load it'll not overspeed and might give its rated power.
I might be wrong but I think this ~ covers the issues mentioned by owen_a earlier.

 

[jim hardy]

I must say, feathering is a lot safer feeling than electric braking but it's a much more complicated mechanism.

Hmmmm don't know why i stumbled back in here.

a clever mechanical engineer could design a blade root that's pulled inward by a spring and engages a spiral track ;
as centrifugal force hurls it outward overcoming the spring it rotates in the helix feathering the blade

automatic, Henry Ford simple, and no computer required .

 

[sophiecentaur]

Centrifugal regulators can be good but not easy to make without a lathe etc.. a bit of electronics can be put together with stuff out of the drawer.
And where am I going to get a clever Mech Eng at this time of night? [emoji4]