Alternatives to batteries unique to my situation

[makenzie71]

You can use this video as an exact explanation of how I'm set up here, but roughly my turbines go to rectifiers, then to batteries, then to grid tie inverters.

I do not use any kind of battery or diversion control.  If my turbines are cranking out 75v, that's what my batteries get.  If the turbines aren't turbines aren't turning then the inverters suck them down to 20v (36v bank).  The batteries are only there to help keep voltage under control, they take a tremendous beating, they are NOT used for storage, and I'm trying to sort out a different way of doing it mainly because my batteries, set up as a 36v assembly, work okay for all my turbines but don't work well for any of them.

I've been told of using capacitors but when I ask for details no one seems to know what capacitors specifically I should use...and that's not an area I know enough about to dive in and try to sort on my own.

I was told I could use resistors...I have actually been told that's essentially what I'm using the batteries as now...but it seems resistors will burn off a considerable amount of my wattage.

What I want is to find a way to arrange things so that I could each turbine to it's own inverter and let it operate at a voltage it likes to operate at.  I'm using cheap ebay grid tie inverters and they really do seem to work okay for tracking voltage when there's some regulation in place, but if the turbine ever gets up above their intended voltage (60v) they shut down and the turbines start free wheeling.

Basically what I'm looking for now is alternatives.  Capacitors and resistors are great, but which ones should I use for, say, a turbine that tends to operate between 20 and 70 vdc and has peaked at over 1000w in the past?  And even now I have a, iSta Breeze Heli 2.0 which has burned way past 1000w at 20mph and that was with the drag of being connected to a 36v bank...it's going to peak somewhere between 2000 and 3000 watts...what should I try with that one?

Ideas and leads would be greatly appreciated!

[bigrockcandymountain]

One thing that comes to mind is a voltage monitoring relay.  It could kick on a big resistor when the voltage spikes. 

What is the voltage range that your grid tied inverters can operate on?

[makenzie71]

The ones I have at present range from 20 to 50 vdc input.

I did try one of the Chinese charge controllers that came with one of my turbines...that's how they operated, if the voltage spiked it'd dump to a large resistor.  The problem I had was that voltage would and basically bring the turbine to an instant stop.

Right now I'm looking at one of those rudimentary dump load controllers like what missouri wind sells...the ones that use the little arduino type controllers with a golf cart solenoid...and basically using my inverters as dump loads, but I'm thinking it'll probably do about the same thing.  Hit the heavy load, stop the turbines, and kill the connection.

[OperaHouse]

I won't say I exactly dump, but I send excess power to water heaters. Unlike relay dumps my dumps are proportional. At a set voltage the diversion starts.  If you want to limit at 75V, it keeps it right there within a few tenths of a volt.  And this is extremely fast response. Your grid tie inverters will easily lock into that voltage instead of searching.

Problem is you have to build it.  All my stuff is custom.  If GHURD still offers a kit this would be very close to it.  You could do it with multiple voltage relays progressively adding more current to hold the voltage down.  In the end these makeshift ways are more trouble than doing it right.

Just to show what I've been doing lately.  This is my third water heater dump load in the garage just to supply hot water to the washer.  This is the lowest priority dump when nothing else needs energy. By coincidence this operates at 60V. It gets up to 129F and can drop down to about 92F after three loads of wash.  The washer runs off panels, no battery. Those finned oil filled heaters are also a pretty good dump too.  Their controls tend to fail and can be found cheap.

[makenzie71]

I'm not looking for a dump solution.  My grid tie inverters are my dump solution.  What i'm looking for is a way to replace my batteries with something that help keep the wind turbines under control and just let the watts go straight to the inverters.

[mab]

Not a simple answer I'm afraid - you have a variable current coming into the gti and need to keep it stable long enough for the gti to match it. A big capacitor would do that but as to how big? That'll depend on the response time of the gti, and the response time of the windmill.

You also need a dump load controller & resistor to limit the voltage on the cap to <50v either because the cap is too small and the gti hasn't had time to respond or because there's more power coming in that the gti can use; but unless you have a HUGE capacitor this controller needs to be fast - as Operahouse says, a Ghurd controller which uses a mosfet may switch on & off fast enough but a big contactor will not.

To give an example: Say your mill is putting out 1kw at 50v 20a and you need the dlc to come on to stop the gti shutting down. If your dump load is, say 2.5ohms,1kw so it can handle the full power of the mill if it has to, then it'll take 20a from the capacitor when the dlc turns on. If your capacitor is 100,000uF then the voltage on the cap will drop at an initial rate of 200v per second (20a/0.1F).if you want to turn the dumpload off at 48v then your dlc needs to turn on then off again in 10ms.

If you don't mind dropping 20v it gets easier(though still too fast for a large contactor), but the gti might not like it.

[OperaHouse]

It always surprised me that there weren't more high speed shunt regulators out there.   Certainly they weren't beating a path to GHURD's door. It is probably all this confusion as to why there is no demand.  That and fear of most things electronic hen it comes to wind protection. The regulator pictured corrects voltage faster than 1,000 times a second preventing over voltage.  That allows me to run the washer off this voltage when the load drops from 600W to 8W without fear of over voltage.  Direct power is almost unheard of presently except for water pumps.

[mab]

Now that you mention it - no there don't seem to be any off the shelf ones.

I tent to make my own so never really looked, but I had thought there were PWM controllers out there - which there are but they're not dump load controllers of course they're regular solar disconnect controllers.

[OperaHouse]

This is the closest I've found that can be modified, a 300W 12V inverter board <$10.  It first came with 75V 75A FET that would work to 70V.  Alas, the Chinese changed them to 55V FET, even changed the chip and created the most horrible trace spacing I have seem.  That makes them more effort to modify than building one.

[richhagen]

Hi, I have a type of setup here that would work perfectly for what you have, well, maybe with a modification or two.  What you need based upon your description, is a buffer, that can handle the voltage of your turbines and yet not be damaged when discharged down to 20V, and also hold enough energy so that your inverters can run smoothly. 

What I have are ultracapacitors.  A note of caution is that you need to keep a limit at the high end voltage that you design for.  Low end voltage is no problem all the way to zero.  Specifically, I have a system at my house which cycles between 60V and 32V depending upon what my situation is with solar and the inverter.  As the voltage comes up loads switch over via relays from grid to solar, and back again on the way down. 

Specifically you need something like the 3000F 2.7V boost caps from Maxwell arranged in a series/parallel configuration that will give you the top end voltage you need and enough capacitance so that your inverter does not excessively cycle.  You may also need a voltage switched dump load should the voltage get to high due to an inverter fault or failure to keep up with the available energy particularly at high winds.  Your wind turbine will output a voltage dependent solely upon its rate of rotation.  That voltage will normally be held down by the batteries or the capacitor bank as those will charge up with the maximum current that your turbine can output. Because the capacitors will have a smaller capacity, there is the possibility that the inverter could fault and stop soaking up energy allowing the voltage to steadily increase, and because there is no way the small loads in the balancing circuits for the capacitors could keep up with this, you would definitely need a voltage controlled dump load capable of using up more energy than your turbine could be expected to put out.  You really should have that on your battery set up if you are putting out a lot of current to keep from murdering your batteries though too.  A circuit to switch a relay or more likely relays to turn on resistance heating would be reasonable I would think.  You just do not want it to fail and over voltage your capacitors.  The capacitors will likely last decades in this role, though from the manufacturer their rated life is ten years although that is at a high cycle rate and adverse temperature conditions.  Some of mine have been in use more than a decade, though I would have to go back and look at when I actually installed them.   It is basically set and forget as there is no real maintenance other than periodically checking the connections.  I don't need the dump load with solar as my controller simply open the circuit when the voltage is too high.  I try to prevent this by progressively switching over more loads until I overwhelm my panels ability to charge up the capacitors. 
 

I built bus bars for mine from 1/8 inch aluminum bar stock and used commercially available balance circuits on them which are the boards that you see in the photo.  Because they cycle much slower in my application I used one balance board setup to cover 4 paralleled capacitors.  A stack of 96 of them 4 parallel and 24 in series stores about 178 Watt hours that I can use.  By the way, I bought mine second hand mostly.  Some of them saw earlier life riding around on hybrid buses.  They are Maxwell brand Boost Caps, I have some rather old ones at 2600F 2.5 volts (stacks with 28 series and 4 Parallel) and mostly 3000F 2.7 volts (stacks with 4 parallel and 24 series) plus a few series with different sizes added to the mix.  The off brands do not appear to be the same thing, so be careful what you pick up.  I would rather have the Maxwell used than some off brand import new, as I had some of those fail to short due to increased self discharge and thus, they are not the same thing in my opinion.  They are a bit pricey for the storage capacity, but I do not spend time fretting about them, they just do their thing once there and I do not think you would need much capacitance with your setup.  At any rate, just my two cents worth.  Rich

[richhagen]

These are NessCap, a major Korean brand of these that has been around a while.  While I have not used that brand personally, I would not have large concerns with doing so.  These look to me to be configured about like what you would need, you would just need to ensure that your voltage never goes above 48V.  You would need to figure out how much energy you need to have available in your inverter's working voltage swing.  The basic math is that capacitors store the energy in Joules equal to one half times the capacitance in Farads multiplied by the square of the voltage.  A Watt is conveniently equal to one Joule per second.  Additionally, when you connect Capacitors in Parallel the capacitance adds, when you connect them in series and they are all of the same capacitance you divide that capacitance by the number in series.  For 18 3000F capacitors in series you therefore would have 166 Farads or so of capacitance.  Say you were running between 48 and 20 volts, one of those setups would give you approximately .5*C*(V1^2-V2^2)/3600 Watt Hours of energy, or 44 Watt Hours of usable energy between those voltages.  If you connected more in series you would add the Watt Hours of each together to form a bigger buffer.

If you are looking at this in the future the link might not be good, but basically the link was to a listing of a bank of 18 3000F capacitors connected in series with integrated balancing circuits for U.S. $315.  I have no connection with the listing or listing person, I just happened to run across it and found it relevant to the discussion.
https://www.ebay.com/itm/Maxwell-Nesscap-UltraCapacitors-2-7V-3000F-Cells-18-Pack-ESHSR-3000C0-002R7B5/254684781143?hash=item3b4c655a57:g:ULIAAOSwk5pfMyZ3