Inverter as dump load

[ruairihev]

Hi,

I have a 24V 1kw wind turbine that I want to use for heating water in a water tank. The turbine is located about 100m from the tank so running 24V is out of the question. I am thinking of using an inverter to send the power from a buffer battery at the bottom of the tower to the tank at 230V, no need for as thick a cable then. I have searched for comments about using an inverter as a dump load and they all say not to do it but don't really say WHY not to do it!

In the water tank I will have three heating elements rated at 250W, 500W and 750W. I have built a controller that monitors the battery voltage and can turn on the elements so that I have outputs of 250W, 500W, 750W, 1000W and 1500W. With the wind I have here the turbine is rarely putting out less than 200W which 24hrs a day every day should at least pre heat my water tank (40 gallons). As a back up incase the inverter fails or shuts down or something else goes wrong I might use a Xantrex C60 or morning star diversion controller at the battery bank to take over and dump load to an air heater or some sort of resistance at the bottom of the tower.

The controller will sense the battery voltage and turn the elements on as needed, I have a bit of hysterises so that they are not rapidly turning on and off. Can I get away with using a modified sine wave inverter (i.e. a cheap one) or will I have to go for a pure sine wave? I think the MSW might not handle the loading and unloading and eventually fry the electronics where as the pure sine might handle better with the toroidal transfomers? I can get four 3000W MSW inverters for the same price as one 1800W pure sine wave but I suppose you get what you pay for. I am hoping that I'd get a good life expectancy from a pure sine wave inverter?

Some might say that I should just change my stator to a higher voltage and feed the elements directly from the turbine, it is an option but I have the stator already built and I also want the inverter for expansion in the future so that if I ever want to run the electricty in the house then I'll have that option by increasing my battery bank.

So, can anyone suggest to me why I should not go with this plan or why it won't work? Is there any other way of going about it that I am missing, using a transformer of some kind?

Thanks for any help,

ruairíhev

 

[ghurd]

"WHY".  Most controllers run way too fast for directly driving a light bulb, let alone an inverter.

It sounds like your controller controls the AC?

Maybe some kind of triac circuit to vary the watts fed to the heating elements.

Not sure what the inverter would think about that.

G-

[ruairihev]

Thanks for the comment ghurd,

Yes, the controller controls the AC side. It basically turns on and off loads as they are needed on the AC side. The inverter will be connected directly to the battery with no fast controller between it and the battery. I will connect a fail safe device (xantrex C60 maybe) in parallel incase of unforseen circumstances.

I suppose its just like using an inverter normally in your home except instead of you manually flicking a switch and turning on a load the control circuit will do it for you!

[Flux]

Yes you can do as you suggest. The inverter will need to run all the time unless you manually shut it down in periods of no wind.

If your controller controls load after the inverter that is ok.

You can cover the case of inverter failure with a back up C60 controller to protect the battery if the inverter fails. Normally the C60 will not dump and it must have its own dump load. There is no way that the C60 can be used to feed the inverter but I think you understand that.

Many have proposed using an inverter after a C60 in place of the dump resistor and that is a definite no go area.

If you want heat alone then this is not a clever scheme as heating controllers work best with turbine volts proportional to wind speed, but it is not much worse than a fixed voltage link at higher volts.

I know absolutely nothing about your turbine, but direct connection to heaters with suitable control may not be out of the question as it would most likely be working well over 24v at times when you are producing significant power. Cut in at 24v would likely give you 70 -80v at full load and line losses may not be that bad.

Flux

[ghurd]

If the only point of the windmill is to heat the water, and the hysteresis is large enough (1V? 1.5V? 2V?), it may work.

If the elements are connected through relays, I don't expect the relays will last too long or be very reliable.  But I don't like relays cycling very fast for long periods.

SSRs would be better.

Might think about an adjustable LVD to shut the inverter down.

Say 3~5 days with no wind with the inverter pulling X-amps, could put a world of hurt on the batteries before the inverter shut down at 21V.

Plus the power the inverter uses during those times needs replaced, and that is power not used heating the water.

G-

[ruairihev]

Thanks for the comments,

I realise now that I shouldn't have titled the post 'Inverter as dump load' but something more like 'Using an inverter to dump load'.

The LVD is a good idea that I will employ, also I think I will use SSR's as the relays will probably not last too long.

I had thought about direct connection but the controller side of it seemed difficult so I said I'd take the easier option in my opinion and use the AC side controlled inverter. If I feel brave in the future then I will try a direct connected heating wind turbine.

I know the answer to this but I need someone to tell me that in the long run its better to buy the pure sine wave inverter than the cheap MSW inverter? Its a difference of €250 versus €1000. But because its only being used for resistive elements the MSW might do the trick?

[ghurd]

I think MSW would be fine for this.

Not sure 3000W is needed is the max load is 1500W.

I would probably go with a 1500W or 1750W.

If the windmill is 1KW, then dumping 1500W would be a rarity I think.

I might even change the 500W element to 250W.  Max load would be 1250W, then a 1500W inverter should handle it.  Here, about 1500W is where inverters start to get expensive faster, in $/W.  The savings may buy a TriStar?

G-

[Flux]

I agree with Ghurd completely. Sine wave inverter would be wasted for resistive load.In most cases the efficiency of pure sine is lower unless you pay a fortune.

Choose your load correctly and a 1500W inverter should do the job easily.

Flux

[frackers]

Sounds like you are going down the same path as I did - and my system is still running even if its a mains voltage heater and not the pool pump yet (gotta run the cables!!)

Check out my logic for this way of doing things here

[Opera House]

Consider using a timer instead of a shut off voltage. Just calculate how many watts you want to dump at a time. I have a freezer/fridge I use as a dump load.  When battery voltage gets to the preset level, it turns on the fridge for 8 minutes.  I use a LM431 as the voltage detect and a MC14541 as the timer chip that can time from seconds to hours.  This turns on a Harbor Freight 2,000/4,000 watt inverter.  These are cheap, costing me less than $20.  Multiple heater ranges is way over thinking it. Operating a 220 heater at 110 just means that it takes four times longer to heat.  Damaged inverters are cheap and there is no reason why you can't just use the DC section if the output FETs are fried. Have multiple dump options and prioritize on just a couple mv difference of trip voltage.  My fridge works great with just this control.  I haven't even bothered to hook up the temperature control I have for it.

[ghurd]

For the LVD,

If the inverter has a regular rocker switch, then a small regular 24V relay with a high ohm coil can be paralleled with the switch.

The LVD will basically turn the switch (relay) off.  No need to deal with large amps in the cables.

(same idea as Speo's 'remote control inverter')

http://www.fieldlines.com/story/2008/4/11/5115/12104

G-