Low voltage immersion connected to wind turbine

[madandy]

HI all.

The more I read about small scale wind turbines and the low overall output the more I am reconsidering buying one.

BUT - Have just found a company that supplies a low voltage immersion (12volt / 250watt & 24 volt / 500 watt) that can be connected directly to the output of a wind turbine (I'm thinking of one of the lower power navitron units). If this would work it seems to me an ideal way to use wind power. Yes the watts are low but if there was a trickle charge over a long period it would heat the water. No batteries, no grid connect system, ? no controller, or dump load just a low power wind turbine directly connected to a resistive load topping up my hot water. Of course cable size would be an issue.

My thoughts are that this system along with my wood boiler and solar panel should cover all of my hot water needs and since according to some people water heating accounts for 65% of the energy used in our homes this would equate to a good way to utilise wind power and save energy.

Any info on this subject would be appreciated.

I have read the previous topics on wind turbine/immersion but they are about mains volatge units.

[wyleu]

Interesting idea, I don't imagine an immersion heater is fusy about it's supply. Lots of isolation probably required and no doubt a raft of labelling but it might well be an interesting way to extend the renewable power input at a time when solar ai'nt doing it's best. Calculating the likelyhood of overheating during windy sunny days might raise one or two issues as you could end up driving the whole lot up to near 100 degrees and that ain't allowed, so you might have to consider somewhere else to dump the extra energy.

[Alan]

Hello Andy

Not sure of the cost involved of these 12 and 24 volt heaters. The cost of a low resistance cable for a twelve volt supply has to be taken into account. Also to keep the alternator load balanced you will have to convert the three phase supply to dc at the alternator. You do not mention if a thermostat is supplied with the heater, if so this will have to be rated for D.C. that question will need to be asked. It will be required if the system is  open vented or sealed. If sealed like a Megaflow you will need a fail safe temperature control  device.
   I have used three double wound transformers here to transfer the power at higher voltage and use the existing  3 K.W. immersion, they were bot that expensive. With the 48 volt alternator this used to feed 1000 w into the existing heater.  Not in use now

Regards

Alan

[madandy]

Thanks for that. Had thought about cable cost and overheating but not about a thermostatic control of any type.

Overheating will hopefully be avoided by linking in the pump on my central heating to dump excess heat (could end up with a very hot house on a windy /sunny day)

The web page is www.wind2watts.co.uk. Can't see any sort of thermostatic device apart from the main cylinder thermostat!

[Ork-NAK]

Alan,

Just picking upon your comment about having tried this, but the system is not in use now.... could you say why not? Any particular lessons to learn from the experience? 

Thanks.

Neil

[johnrae]

Anyone out there interested in hearing about my experiences in doing this very thing - 1kw unit tied directly to an immersion heater
Is so just nod and I'll put pen to paper - or fingers to the keyboard
Jack

[roys]

Get those fingers tapping .  I would like to read about how you get on.

Cheers Steve

[camillitech]

Anyone out there interested in hearing about my experiences in doing this very thing - 1kw unit tied directly to an immersion heater
Is so just nod and I'll put pen to paper - or fingers to the keyboard
Jack

me too matey, been thinking along these lines myself

[johnrae]

OK then, here goes

Some time ago I bought a 1KW 48v unit from Futurenerergy - saw it on e-bay long before I found this site - with a view to feeding an immersion heater in my large hot water tank.

Because I wanted the raw 48v 3phase to permit voltage manipulation I had to manufacture my own slipring assembly to get the three phases down the mast - they only offered a 2 way slipring on the basis that they rectified the AC at the generator housing.  I made this from bronze and tufnol.

I then thought of how to get the 48v 3phase 1KW suitable matched to the existing 3kw/240V immersion heater in my water tank.  On the basis that I'd only 1KW to work with I needed to supply the 3KW unit with somewhere around 140v.

On the basis that I'd need to raise the 48v to 140v I needed a means of transforming a 3phase supply into a single phase at higher voltage.  The way I did this was to buy three toroidal transformers, each rated at 500VA, with a stepdown ratio to convert 230 to 70.  Used in reverse these would raise 48v up to 155v.  Allowing for a bit of voltage losses, this seemed reasonable.

The transformers were connected in delta configuration on both LV and HV sides.  The HV sides were then connected via three bridge rectifier units to produce a DC suitable for feeding to the immersion heater.

To monitor what was actually being fed to the heater I coupled up a Lascar data logger, type EL-USB-3.  This measures DC voltages (0-30) and with a suitable potential divider system it was set to measure 0-200V.  These can log 32000 readings so at 1/minute I could get 3 weeks of logging before I needed to do a data dump.  Note that 1/min for a month needs 44,000 lines in an Excel spread sheet which is beyond the graphing capability of Excel.  So much for Bill Gates and his expensive software.

Overall, results were very disappointing with very low output, but the odd occassional burst up to 1KW in high winds.

I came to realise that the extremely low DC resistance of the transformer windings were acting as an electrical brake and hence preventing the turbine from rotating in anything but a good blow.

So out went the transformers and onto the MKII version

I then bought a 48V 1KW immersion heater and fitted this in place of the 240v 3KW unit.

Again the 3 phases were rectified and the voltage monitor adjusted to give a full scale logged value of 60v.

Results were now better but again not up to expectation.

The resistance of a 48v 1kw heater is only 2.3 ohms and I believe that, because this was permanently connected onto the turbine was again acting as a brake and preventing rotation in all but a good blow.

Even my neighbors were commenting on how rarely they saw it rotate - very embarassing !!

Now for the MKIII version

So thinks I, what I need is a means of disconnecting the turbine from the load until such time as it was rolling nicely then apply the load.
What I did was build a switch circuit which would close once the turbine was running at about 40% speed and then stay closed until it stopped rotating, at which point the switch would open and then remain open until the turbine again achieved 40% speed.

This was achieved by using a large power thyristor in series with the rectified feed to the heater unit.  The thyristor itself is triggered by a simply capacitor discharge circuit which is triggered by the rising DC level, set to operate at 40 volts.  Once the thyristor is turned "on" it remains conducting until the supply voltage has dropped to less than 1 volt.  At this time the capacitor trigger pulse is achieved via a relay.  It won't last forever but at least it proves the concept.

What now happens is that the turbine runs up in even the slightest breeze, reaches 40 volts and is then connected to the heater unit.  If the wind energy is insufficient to maintain the load the turbine slows to a stop.  If the wind energy is high enough then the load is sustained.  In effect the system is being pulse fed in light winds with the inertia of the turbine acting as a mechanical capacitor.

At least my neighbours can now see the fan spinning, even if it doesn't do anything until it reaches the majic 40% value.  In other words I only get power if there's more than 160watts available.

All this is less than ideal but it does at least work and I'm finding that my logged data is now showing more power to the heater.

Ideally I should be modulation controlling the three phases via triacs such that they may be regulated so that load power exactly matches wind power with the turbine running at optimum speed.  But that requires some thought - unless someone out there can come up with a suitable circuit.

Circuit and/or photos (not very pretty) available if required

regards
Jack

[Ivan]

Hi Jack,

I'd pretty much discovered what you have discovered. The only success I had was running a heater element (about 700-1000W) from one phase of a 2kW wind turbine. Not to be recommended, as it would easily damage a turbine. I don't imagine it would last long if permanently connected.

I think you are on the right track - try to do what a battery does - batteries provide an almost ideal load. You can try to be cleverer than a battery but it's easy to get it wrong. I suggest (take a Navitron 48v 1kW wind turbine as an example):

below 48v - leave disconnected
At 48v, switch in load 1 (say 300W)
At 54v switch in additional load 2 (say 500W)
At 56v switch in additional load 3 (say 1kW)

You could make the voltage points adjustable, which would allow you to tweak the unit in operation.

This is the best way to do it. If you have time, budget, you could have an extra stage or two of load.


Ivan

[IainB]

Is it  worthwhile having a battery bank and charge controller if all that is required is resistive heating?

I appreciate that of the turbine and will start at a higher wind speed, but from Weibull,  low wind speeds are infrequent and contain very little power anyway.
Mean windspeed here is 6.7 m/s at 10m, I was considering the 1kW model connected directly to  2x 24V 600W immersion heaters in series.

I tried changing the cut in speed on the calculator:

www.reuk.co.uk/Calculate-kWh-Generated-by-Wind-Turbine.htm

3 m/s - 7630kWh
4m/s - 7565 kWh
5m/s - 7415 kWh
a difference of £21/year

What would the cut in speed be in real life??

[johnrae]

Ian

Be aware that the 600watt units have a resistance of only 0.96 ohms so when tied in series this presents the turbine with a load of 1.92 ohms. 

This will in turn apply quite a braking effect to prevent turbine starting in anything but a good breeze

See my notes re disconnecting the load until the turbine has reached a decent speed.

When you consider the ideal load characteristic you find it's infinite resistance at zero speed , decreasing exponentially to a minimum resistance (maximum load) at full rated speed. 

The simplified equation for calculating load resistance is  R=k/Wind Speed squared where k is a constant determined by turbine capacity at a particular speed. 

For a 48v 1kw unit rated at 12M/s then k = 331

regards

Jack

[IainB]

Jac

Ta for that

<BTW there are two "i"s in Iain : )
"There's no "I" in team........though there is a "me" if you mix the letters up ........." >

From P=IV and R=V/I, I agree with your 0.96 Ohms for the 24V, 600W elements. 1.92 Ohms total.

From your equation: Windspeed = squarerootof(331/1.92)  I get 13.1m/s

Have I understood that correctly? Is that the new (higher) cut in windspeed? it seems too much.

I saw a graph of RPM vs voltage for a PM generator - it started at 0,0 and went up in a straight line, Volts proportional to RPM. Power OP would also be linear from P=IV, Resistance being nearly constant.

That being the case, I expected to have to only overcome the static (sticky) permanent magnetism initially, then the resistive load would absorb the power progressively.

I understand also that the aerodynamics are ineficcient for a stalled rotor (from Hugh Piggot's book)

In the end you have the real life experience - Mk II was "better" , but still needing a "good blow" - can you estimate the startup windspeed for MkII? - I'll plug it into the calculator and see what difference it makes to the annual production.

Iain

[johnrae]

Iain,

Looks like I've totally screwed up on my equations.  So much from working from memory

Load R is inversely proportional to wind speed Sw  or  R = k/Sw

In simple terms for a 48v 1.2Kw (your 2 times 600watt load) rated at 12M/S    1.92 = k/12    ie   k = 23

Apologies for that major foopah

Haven't determined what the exact cut-in speed is for my system but it breaks away fairly easily now and at 40% speed the full load is applied.  This is less than ideal and I need to work out how to effectively apply a "progressive" load using only one fixed value of actual load.  The impossible just takes a little longer !!

Jack

[john macbeth]

Hi Guys, for those of you with low voltage systems, you may want to have a closer look at this Siemens PWM. RS Stock No.   469-3167.
£25 plus VAT.

I came across this last week in search for high voltage PWM.

PWM's connected to heating elements can be very noisy, you might find you need some large capacitors before and after it to keep it quite. The alternator on the turbine may also suffer from some harmonics at certain chopper frequencies. 

The control is a o to 10 VDC input corresponding to 0 to max at the load output. I think you can also program a cut in point too, for your wind turbine cut in speed. Can not remember now.

So all you need is a potential divider after your rectifier to get an appropriate input signal. Probably a 10V zener would be worth wile putting into the circuit too.

With a bit of soldering and a few resistors you could have a very nice MPP tracker.

Hope this can work.

Cheers


John