High Voltage Wind Turbine

[Sparky01]

Hello all. I have been hanging in the shadows taking it all in for a while and have put up a couple of turbines fairly close to my house (but there are turbulance) to date. But I seem to have an issue that I do not see too many postings about. I would like to know which would be the best to with. My system is 12VDC and my ideal location for a turbine is 350' from my battery room.

I also would like to have room at this location for muliple 10' turbines and testing.

Option #1 Build 120VAC wind turbines so I can use small wire and use a transformer and rectifier at the battery room (I allready have transformers)

Option #2 Shell out some big $ for huge wire to be able to handle the amount of amperage that I would be looking at. I realize that the Transformer setup will loose about 20% eff. that will hurt on the low end. But wire losses @ 12v will hurt on the high end.  Also is it worth the trouble installing a Delta/Wye switch using RPM. If so does anyone have a link.  Thank you all

[Nando]

The transformer do not have 20 % loss, around 3 to 5 % is normal so you will get 95 % efficiency plus the 350 feet about 3 to 4 % for a total of 90 %.

Low voltages wind mill end up being around 50 % efficient ( broad range), then you need to add transmission losses.

High voltage if the load is 9 times the generator impedance the efficiency will be 90 % plus the 90 % transmission and conversion for a total of 81 %.

Generator producing 380 volts at 12.5 m/s is ideal for conversion to lower voltages

At lower RPM you may obtain 220 volts,

If more info is needed, just ask

Nando

[DingBat]

An option I've given some thought to is to use a pulse charging circuit like Jerry uses to charge his electric truck.

Downsides of using pulse charging:

Reduced blade loading (how much ?)

Possible vibration caused by pulsing load

More wires in run from turbine to batteries

Cost, might need a lot of run capacitor to be workable.

Upsides:

Reliability, all passive components

Bridges near batteries, easier replacement

Diode drop automatically corrected

I'm still not quite sure how the frequency, amperage and capacitance relate for such a design. See the 24 uF note in the diagram of wooferhound's comment.

-- DingBat

[Flux]

You will get a lot of mixed comments about this one.

If you intend to use multiple 10ft turbines my advice would be to abandon the 12v system and move up to at least 24v and better 48v, but I suspect you will not do this.

There are many options, some better than others, none are cheap.

Direct transmission at 12v is not cost effective with new wire, but at 12v you can use any old scrap cable and parallel it up as long as you make sound connections the insulation requirements are negligible.

Transformers are perfectly satisfactory if you do it right, it is easy to fall into a lot of traps.

Single phase is not really viable so don't do it. use a 3 phase transformer or 3 single phase ones.

These need to handle the frequency of your alternator and generally it is very low, making the transformers large and costly for even reasonable efficiency. it would make a lot of sense to build alternators with many more poles than standard to raise the frequency.

The best method is dc to dc voltage converters but it is a specialist field and may not be an option for you.The one possibility is that you may be able to modify computer power supplies to do the converting, they are cheap and fairly tough but again it is experimental ground.

Which option to choose depends on many factors, the thing is to understand the implications of each scheme and choose what suits your needs best.

Flux

[Sparky01]

The X-formers that I have are 240VAC input, 24VAC outut. So 120VAC from the turbine= 12VAC out. also will the Hz be within reason for the X-former? This is from the 3PH 12 pole stator. 150RPM cut in = 30Hz. as far as the high end RPM I am not sure what the Hz would be.

[Sparky01]

I forgot to say that yes the transformers are 3Phase. It came from an industrial power supply. I also have the 3 Ph rectifiers that went with them. I have a question on the rectifiers is they have large capacitors in paralell with the output. Should I remove these? Also should I have around 350 turns per coil to get the desired ouitput?  Thanks for the help

[Flux]

No, its 15 Hz at 150 rpm.

Flux

[Sparky01]

The formula is not 150 rpm x 12 poles / 60 sec. ?

[Flux]

The minimum number of poles you can have in a machine is 2 and that gives 1 cycle per rev. 12 poles will give 6 cycles per rev.

150 rpm gives 150 x 6 /60 Hz = 15 Hz.

I never use a formula I just remember 3000 rpm is 50 Hz for a 2 pole machine and work from there.  You may prefer to remember 3600 is 60 hz if not in Europe.

This does make your transformer rather borderline. It will have a high magnetising current and it is likely that the mill will not start with the transformer connected in low winds.

Now you see why I suggested using more poles. The transformer will be reduced to 1/4 of its 60 Hz rating in terms of VA. but you can still use it for the same current as the 60 Hz rating.

Flux

[Ungrounded Lightning Rod]

When charging batteries, if the rectifier is a bridge, the capacitors are redundant.  You can take 'em out or leave 'em in at your convenience.

Leaving 'em in raises the risk of failure, since a cap could blow - especially if you come unhooked from your batteries in a high wind.  Taking 'em out is extra work.

[Ungrounded Lightning Rod]

When charging batteries, if the rectifier is a bridge, the capacitors are redundant.

Assuming they're DOWNstream of the rectifier.  (If they're upstream something else is going on - like a regulating transformer that depends on a tightly controlled frequency.)

Probably not an issue since this is rectified so they're probably output filter caps.

[Sparky01]

The capacitors are on the DC output side

[Ungrounded Lightning Rod]

The transformer will be reduced to 1/4 of its 60 Hz rating in terms of VA. but you can still use it for the same current as the 60 Hz rating.

However it may saturate if you run it near the rated VOLTAGE at a massively lower frequency.

If you want to run 120V -> 12V at 15 Hz using a 60 Hz transformer, use one that is rated for 480V -> 48V.

(If you match it to the generator at one speed, by the way, it will automatically be matched at all speeds.  A generator is just a secondary-only transformer with rotating magnets replacing the primary.  The field from the magnets is constant, so the induced field in the transformer is essentially a constant amplitude - dropping off at really low frequencies due to resistive losses.)

[Sparky01]

I geuss that I can find the schematic of the machine that these came out of( it was a VCR tape assembing machine). I was just wondering if the capacitors work well to keep the sine wave constant or do I want the high spikes?

[Ungrounded Lightning Rod]

The X-formers that I have are 240VAC input, 24VAC outut. So 120VAC from the turbine= 12VAC out.

Yep.

also will the Hz be within reason for the X-former? This is from the 3PH 12 pole stator. 150RPM cut in = 30Hz.

If that were true (30Hz at cutin) you'd have done the right thing.  You'be be the rated voltage at the rated frequency and half the rated voltage at half the rated frequency, which is correct for getting the most out of your transformers while avoiding core saturation.

But 12 poles is 6 cycles/turn, not 12.  So you're 15 Hz at cutin.  (As has been pointed out elsewhere.)  So you'd have to wind your genny for a cutin voltage of 60v and change the ratio to avoid core saturation.

Can you change the taps, or stack two transformers per phase, to run them as 480 -> 48?  Then you could wind the genny for 120V at cutin and be fine.  Or could you build a 24-pole 120v genny to use them as is?

as far as the high end RPM I am not sure what the Hz would be.

Doesn't matter.  The transformer core will have the same magnetization (within a small delta) at any frequency, because the voltage (and thus the current ramp rate) increases with frequency, leading to constant amplitude reactive current.  I think eddy current losses will be frequency-independent as well.

Other than core saturation your main transformer limit will be heating, which is mainly I-square-R losses from resistive heating.  So keep your currents no higher than the rating and you're fine.

Doing that consists of getting your furling adjusted right.  It's pretty much like using furling to protect the alternator, except that the transformers don't get extra wind cooling at higher speeds so you have to pay attention to the actual current rating - rather than going somewhat higher as you could on an induction-motor conversion.  (Of course if the transformers are oversized for your alternator you just have to adjust furling to protect the weaker device.)

[commanda]

Have a look through my diary.

I've played with the Fisher & Paykel with 3 x toroidal transformers.

There's also a circuit of a tacho with star-delta switch.

The actual transformer specs (primary & secondary volts) tend to be misleading, since once it reaches cut-in it becomes more current-mode than voltage mode.

Best bet, if you can, is set the generator up in a drill press and test the transformer setup that way.

Amanda

[Flux]

The capacitors were for smoothing in the original application, the battery will act as a capacitor in your case, they can stay or they can go, they will add no benefit.

With a wind generator clamped to a fixed battery voltage, there are no sine waves, all the voltage and current waveforms will be chopped badly.

The clamped voltage waveform will make the transformer magnetising current worse at cut in, but because the volts are clamped the flux density will fall as wind speed picks up. this is not the same as a resistive loaded case where you you operate at constant volts per cycle.

your transformer will be saturated at cut in, you will not damage it as the power source is current limited but it will make starting very difficult and low load efficiency poor.

[Sparky01]

I am not sure that I understand how a different voltage input transformer will help with the input Hz problem

[kitno455]

as i understand it, it will be bigger, so is less likely to be saturated at your cut-in rpm. but i dont know much about this

allan

[Flux]

For a given size of core, the flux density is dependent on the frequency and the number of turns. The voltage rating of a transformer decides how many turns it needs to avoid saturation. A high voltage will have more turns.

If you halve the operating frequency you need twice the turns.This is the same as using one rated for double the voltage at the original frequency.

Now you can see why I said that transformers are satisfactory if you know what you are doing, you have the sense to ask, I suspect others mess it up when they try and blame the system.

Ideally for wind use you should operate with lower flux density to reduce the magnetising current and help start up. This de-rates the transformer even more.

I suspect some miss this point when I say that if you have to buy the transformers they will seriously increase the cost of the system.

Flux

[Tom in NH]

This is a totally hairbrained idea, but I'm curious what other more experienced folks have to say about it. What would happen if you rectified your 12vac at the wind turbine and fed it into an el cheapo inverter to convert it to 120vac which you then either run to your battery shack or transform it to even a higher voltage.

You'd probably need a regulator to keep the rectified dc to within the tolerances of the inverter. I know most inverters have an undervoltage shutdown that may require that you push a button to reset it, if it should occur. I wonder, if one was to bypass the undervoltage shutdown, what effect would that have on the inverter output and on the inverter itself? --tom

[Sparky01]

Ok it is all starting to make sense now(thanks for some great expanations)  So far I think that I will use a 480V - 120V 3Ph transformer( they are esier to find than what I have now) and install it in my battery room. If overheating may be a problem I can install a temperature operated fan for cooling. I am also going to run a CAT5 cable from the turbine to monitor RPM (in a nice warm spot) and was wondering instaed of using a Delta-Wye relay at the turbine can I do this on the transformer instead(i would think that it would be doing the same thing. Would someone knowledgable here let me know if they think this system will end up being overly complex and inefficent(I Build and install machinery automation circuits and tend to overcomplicate things a little  Although so far this seems to be too simple of a solution to many peoples long cable problem. (a used 3Ph transformer = 100-200 Dollars) Also can I have too large of a transformer? I have access to 5KVA, 15KVA, 30KVA or 100KVA 480VAC primary 240VAC multitap Transformers.

Great input so far. THX

[Slingshot]

I think the big problem would be finding a transformer to work over the broad range of frequency present in such a system.  At what would normally be startup, you might be asking that transformer to work at 10 Hz, depending on the alternator.

[Ungrounded Lightning Rod]

I think that I will use a 480V - 120V 3Ph transformer

Running at 48 from the genny (or 83 if you hook the secondary in delta) to your 12V batteries?

You could also double that by using two transformers, putting the primaries in series and secondaries in parallel.

I ... was wondering instaed of using a Delta-Wye relay at the turbine can I do this on the transformer instead(i would think that it would be doing the same thing....)

Yep.

In fact it's better to do it after the transformer.  That way things stay balanced between the transformer and the generator.

Also can I have too large of a transformer? I have access to 5KVA, 15KVA, 30KVA or 100KVA 480VAC primary 240VAC multitap Transformers.

Yes.  Transformers take some power - and some "reactive current"  (90 degrees lagging the voltage) - to stay magnetized even when not under load and the bigger the transformer the more they take.  I-squared-R heating in the genny, transformer, and wiring between them, depends only on the current and doesn't care about the phase, so it represents real energy lost which must be made up by generation, and thus puts a load on the blades.  So an oversized transformer wastes more power.

(Someone has also posted saying that transformer current represents a load that hampers mill startup.  I have no direct knowlege about that.)

You want to size your transformers for the load with a moderate safety factor and leave it at that.  Current rating somewhat exceeding the current you'll put thorugh it, KVA rating somewhat exceeding the KVA you'll pull at peak power times 60 Hz / freq-at-peak-power.

[Ungrounded Lightning Rod]

(KVA rating is the one that shouldn't be excessively oversize.)

[commanda]

To monitor rpm, pull a tap off the ac side of your bridge rectifier. Only need to run control cable back to the tower base to operate the star-delta relays.

Tach circuit & star-delta switch is in my diary.

http://www.fieldlines.com/user/commanda/diary

Amanda

[hvirtane]

Has anybody used this

pulse charging system

with her/his high voltage

wind machine?

In my opinion it is

an interesting option.

I've used a similar system

to charge batteries

from the grid.

My experiences are positive.

- Hannu

[Nando]

Let's clarify some transformer ideas.

A transformer has an inductance produced by the primary of it, and this value if we analyze represents about 3 to 5 % of the energy into the transformer.

Let's say that we have a 1 KW transformer to be attached to a wind mill producing 240 volts at its peak RPM ( like for 12.5 m/s)and 60 HZ output.

Let's assume that it has 3 % losses due to the winding inductance --I am not considering the Resistive losses --.

The inductance of the transformer represents in this case 30 watts RMS with peak power of 30 * 1.41 = 42.3

We use the standard formula for an inductor V= L*di / dt

V= 240 * 1.41= 338 volts

dI = 42.3 / 338 amps = 125.15E^-3 amps

dt = 1 / 60 * 4 = 1 / 240 = 4.17 milliseconds

We use the peak of the sine wave to find the inductance value

V= L * di / dt ; L = V * dT / DI ;

L = ( 338 * 4.17E^-3)* 338 / 42.3 = 11.26 Henrys

Let's say that the wind mills tarted to produce 1/2 of the voltage and 1/2 the frequency, logical since the RPM has dropped to 1/2

V1= 120 Vrms = 169 Vpeak

Frequency = 30 hz ; peak time = 1 /(30 * 4) = 1 /120 = 8.33E^-3 seconds

Let's calculate the inductance current

V= L * dI / dT ; dI = V * dT / L= 169*8.3E^-3 / 11.26 = 125E^-3 amps

SAME AS BEFORE

The transformer under a wind mill voltage generation with variable input voltage and frequency defined by the wind mill parameters will have the same power losses within certain broad operating range.

The only consideration left is to define the resistive value of the transformer to allow the wind mill to cut in.

The next step is to define at what input voltage the system will start using the energy and this requires a decision for transformer conversion via tap setting of the output winding if full isolation is required.

Another way is to use 1, 2 or 3 transformer in series with selective shorting of transformer inputs and parallel rectified output to obtain the desired output voltage.

I have done 3 transformers in series for 380 volts wind mills with output to charge 48 volts batteries .( 9 transformers for 3 phase)

Or 3 transformer with taps ( 3 phase wind mill) to obtain the desired output for 120 volts batteries.

Nando