Talk to me about open voltage

[Astro]

So as I have been getting my parts list together I do not know what size charge controller to get. I understand that I will find all this out when I test with the mill, but just wondering what you guys have seen with your set ups.
I am winding for a 24v system. I understand rpms dictate as do other factors, but I am just wondering what others have seen for open volt numbers. I realize that between batteries and a dump load it will never see the open volts, but I am just curious. Or does a controller see that open voltage while switching from charge to dump?

[kitestrings]

I don't recall the size you are contemplating, or if you've determined this Astro.  Generally there are two approaches - there are more, but these seem most common - one, is where you are tied to the nominal battery voltage, and you divert, or burn off excessive available power.  I think this works well for small to moderate size systems.  I think the Tri-star and Xantrex controllers are some of the more common commercial ones used.  Gurd, and others have made their own.

The other is to use a MPPT controller, like the Midnite Classic (in Wind Mode), and there are some comparable grid-tied inverters (e.g. Aurora).  This is what we have.  Here, the input voltage needs a bit of 'headroom' at cut-in, but then can fluctuate much higher than the battery voltage. The controller, bucks the voltage, as it attempts to follow a user-input power curve.

In our case our nominal voltage is 48V.  We cut-in at about 58V, and the voltage can go up to 150-160V, but we try to live well below that level (110-120V).

With a direct-tie, I think Hugh recommends shooting for a cut-in of roughly nominal plus 1.4V on a 12V system (2x on 24V, right).  I would think wind for something near that mark.  The forcing volts will be linear above that point, rising 2-2.5x the cut-in rpm (before helicopter noises commence) and the challenge becomes being able to divert all of the available power from the thing once the batteries are happy, and or start to furl, or otherwise govern.

Hopefully you never run the thing unloaded, or truly open-voltage, as this is where things get hairy in very short order, but I don't think that is what you meant.

[Astro]

I don't recall the size you are contemplating, or if you've determined this Astro.  Generally there are two approaches - there are more, but these seem most common - one, is where you are tied to the nominal battery voltage, and you divert, or burn off excessive available power.  I think this works well for small to moderate size systems.  I think the Tri-star and Xantrex controllers are some of the more common commercial ones used.  Gurd, and others have made their own.

The other is to use a MPPT controller, like the Midnite Classic (in Wind Mode), and there are some comparable grid-tied inverters (e.g. Aurora).  This is what we have.  Here, the input voltage needs a bit of 'headroom' at cut-in, but then can fluctuate much higher than the battery voltage. The controller, bucks the voltage, as it attempts to follow a user-input power curve.

In our case our nominal voltage is 48V.  We cut-in at about 58V, and the voltage can go up to 150-160V, but we try to live well below that level (110-120V).

With a direct-tie, I think Hugh recommends shooting for a cut-in of roughly nominal plus 1.4V on a 12V system (2x on 24V, right).  I would think wind for something near that mark.  The forcing volts will be linear above that point, rising 2-2.5x the cut-in rpm (before helicopter noises commence) and the challenge becomes being able to divert all of the available power from the thing once the batteries are happy, and or start to furl, or otherwise govern.

Hopefully you never run the thing unloaded, or truly open-voltage, as this is where things get hairy in very short order, but I don't think that is what you meant.

I was thinking a Xantrex. I am just getting going on the whole project and it seems to fit the bill for what I want/need to start with.
Thank you for the information about recommended cut in speed and what voltage you are seeing. I am going to build a vawt, so a few things might be different, but it gives me a ball park and that is what I was looking for.
I honestly probably do not need a 60 amp C60 controller, but I like the higher capacity of open volts it provides.
Plus it allows me to grow in the future if I decide to build a bigger mill and then I can buy a smaller controller for this mill then.

[Astro]

Little controller have an open voltage of 50-60v and bigger ones above 100, that is why I was asking. I am wondering if it ever really sees the open voltage anyway if it switching between charging and dump load?? But also I was thinking about buying one of the cheapest charge controllers I could find, just to play with in the shop. It also occurred to me that a guy may have a bigger controller on a bigger mill and a little one on a little mill and I thought I would ask about them. Like I said, do they really ever see the open voltage if they go from charging a bank to dump load?

[Adriaan Kragten]

On my website: www.kdwindturbines you find a 200 W battery charge controller including a voltage controller and a dump load at the bottom of the menu KD-reports. The dump load contains one transistor and two resistors. This battery charge controller is meant for a 24 V lead acid battery. The wind turbine has a 3-phase PM-generator and the current is rectified. At low wind speeds, the open DC voltage after the rectifier is lower than the battery voltage and so the voltage controller gives no base current to the transistor on the dump load. This transistor is therefore open and no current is flowing through the dump load. If the open voltage reaches the battery voltage, current will flow into the batteries. The loaded voltage depends on the current and on the charging state of the battery. If the loaded voltage becomes higher than 27.6 V (can be adjusted), the voltage controller will gradually close the transistor on the dump load and so a part of the generated power is dissipated in transistor and the resistors of the dump load. The voltage controller is that sensible that the maximum voltage is maintained at 27.6 V independent of the current.

So the voltage controller of my system feels the open voltage only for low wind speeds. However, I think that this is the case for any system. The open voltage increases linear to the rotational speed and at high currents, the open voltage can be a factor three higher than the loaded voltage. So if the load is disconnected at high rotational speeds, the voltage will peak and a high peak voltage may destroy certain electronic components on the voltage controller. So any system must be designed such that the dump load is always working in parallel to the battery and that there is no sudden switch from the battery to the dump load and certainly not a switch such that you have the open voltage in between. So for a well designed system, the voltage is maintained at the maximum charging voltage which is allowed for the chosen battery. The ratio in between the power which is going into the battery and the power which is going into the dump if the maximum charging voltage is reached, depends on the charging state of the battery. If the battery is almost full only a little power goes to the battery and almost all power is dissipated in the dump load.

[Astro]

On my website: www.kdwindturbines you find a 200 W battery charge controller including a voltage controller and a dump load at the bottom of the menu KD-reports. The dump load contains one transistor and two resistors. This battery charge controller is meant for a 24 V lead acid battery. The wind turbine has a 3-phase PM-generator and the current is rectified. At low wind speeds, the open DC voltage after the rectifier is lower than the battery voltage and so the voltage controller gives no base current to the transistor on the dump load. This transistor is therefore open and no current is flowing through the dump load. If the open voltage reaches the battery voltage, current will flow into the batteries. The loaded voltage depends on the current and on the charging state of the battery. If the loaded voltage becomes higher than 27.6 V (can be adjusted), the voltage controller will gradually close the transistor on the dump load and so a part of the generated power is dissipated in transistor and the resistors of the dump load. The voltage controller is that sensible that the maximum voltage is maintained at 27.6 V independent of the current.

So the voltage controller of my system feels the open voltage only for low wind speeds. However, I think that this is the case for any system. The open voltage increases linear to the rotational speed and at high currents, the open voltage can be a factor three higher than the loaded voltage. So if the load is disconnected at high rotational speeds, the voltage will peak and a high peak voltage may destroy certain electronic components on the voltage controller. So any system must designed such that the dump load is always working in parallel to the battery and that there is no sudden switch from the battery to the dump load and certainly not a switch such that you have the open voltage in between. So for a well designed system, the voltage is maintained at the maximum charging voltage which is allowed for the chosen battery. The ratio in between the power which is going into the battery and the power which is going into the dump if the maximum charging voltage is reached, depends on the charging state of the battery. If the battery is almost full only a little power goes to the battery and almost all power is dissipated in the dump load.

That is what I wanted to know. Thank you. It makes sense now. I don't know why but I was thinking the switching action was more like a relay and it was either on or off.
So when playing with or choosing a controller, open voltage limit is not all that important. The ability to handle the amps are, but open voltage is not, because it will never truly be opened up like that and will always have a load on it, even when switching between battery and dump load.

[Astro]

Ok now I have another question.
So we rectify our 3 phase to make dc voltage. Do we rectify at the mill, or in the house? At the mill means low voltage will be running thru the buried cables, and they won't have to be as deeply buried and you will have one less wire. However low volts and high amps means larger wires.
Now if we are using a heater of some kind as a resistor on our dump load, would it be possible to make a system that takes the rectified voltage and charges the batteries and when it goes to dump load, we use the ac voltage from the mill for the heater resistance load? Seems to me that higher voltage = lower amps and thus the ac open voltage coming in would be better suited and more efficient at making heat.
I think this is very possible unless I am missing some aspect of it all. I am unsure how you would keep the charge controller powered so it knows when to switch off dump load, but there has to be a way. You also might be better served with a variable resistance load so that as the voltage rises, you can add more resistance ( up the amps) to bring the volts back down. You would then never need to turn your mill out of the wind, because the faster it spins you have another controller that just keeps adding resistance to slow it back down. You could accomplish the same thing by having multiple heating elements of ascending resistance controlled by different relays. As the voltage goes up you fire the next heater element in line. As the voltage goes down, it starts dropping elements off. Instead of going all or nothing dump load into a very high resistance load.
Now, there are guys on here that make me look like a retard when it comes to electronics, but without thinking about it much (still on my first cup of coffee) I think I could come close to such a thing with relays, soliniods and such. So I am betting if I can make it big, you all know how to put it on a board and make it small.
Don't you think this is possible?

I should add that I spent tons of time working on ww2 type equipment. It was nothing to open a cabinet and see 200 relays. So I understand relay logic very well. I did have some training in plc control and understand it a little. In fact some of my work was to update some old equipment and get rid of some of the relays in exchange for more modern controls. That training was many years ago and pretty basic stuff. To me it was taking the prints of relay logic and putting them into a computer program. I guess my education is in a deeper understanding of how and why relay logic works and less in electronical means of accomplishing the same thing. So I will often talk about my what ifs and how abouts in relay logic and if you try to explain it to me in electronic terms, please understand that I do not know the terminology. Not that I am not willing to learn another language, but that there is going to be a learning curve and maybe some mis communication in expression of ideas.

[Astro]

I think we need a controller that
#1 can fire a relay for an automatic transfer switch so that should we ever have our system run low and need grid power, that it does not switch back to off grid until the batteries are fully charged again. That should extend the life of the batteries.
#2 we need a controller that has its own rectifier and stays operational at all volts the mill is putting out.
#3 we need a controller that has 3, maybe 4 relays to power dump loads of the ac voltage the mill is making and that fires them and drops them back off in sequence and does so at X voltage. X voltage would need to be adjustable, because all mills are not created equal.
 Then all we have to do is use a smaller resistance heater. Well several of them actually.

 Such a system would take advantage of every volt/amp. It would also act as a breaking system, while doing so as a soft break and not a slamming on the brakes. You could with enough relays and small resistance heaters make it so that it is still going to turn the same speed in 80mph winds as it does in 20mph winds. Well, within reason, because eventually you would come to the point of adding resistance that equals a direct shorting.

[DamonHD]

"Fully charged" may be harder to define then you think, and you may also waste a lot of energy that the battery cannot absorb as it gets nearly full.

Rgds

Damon

[bigrockcandymountain]

You have a pretty good handle on what you are up against.  To me, the simplest and best is still to tie the mill directly to the battery with a rectifier between. Then the dump controller just removes excess power from the batteries to keep them happy.  A battery will even out a lot of surges from quick wind gusts. You can think of this as no wind turbine controller, just a charge controller that manages the battery. You can also direct connect solar, hydro, dc generator, wind etc and the one controller will keep the battery happy.   

 I like to have the rectifier in the house by the battery.  Then you can access the 3 phase side.  I have a secondary dump load on mine that has 3 elements in delta configuration that is triggered by high voltage on the turbine.  It acts as a power limiter in that it burns off the extra power before it gets to the rectifier or anything beyond. It heats water.  It also adds redundancy.  I used a 3 phase solid state relay, but mechanical would work fine. 

A variable dump load with different resistances would be great, but it gets a little complicated and to me isn't worth it.  The very high wind days we have so much power that wasting a bit from the turbine being clamped to a lower voltage is fine. 

These are great things to have straight in your mind before you start, but it will all be clearer when you have the turbine up running and can see exactly how it acts.  They all seem to have slightly different behavior.

[Astro]

"Fully charged" may be harder to define then you think, and you may also waste a lot of energy that the battery cannot absorb as it gets nearly full.

Rgds

Damon

In a relay system yes because it is absolute. But in a computer controlled or a electronic system, no. Because as it starts needing 3 amps or 10 amps for charging, it would make up that power by dropping off the relays to the resistance loads. You might be charging and have one or two resistance loads going, you might be charging hard and have no resistance loads going or you might not be charging at all or very little and have several resistance loads going.
With relays you would have to monitor the incoming voltage after the rectifier and before the charge controller and vary your resistance loads by that number. So if you wanted say 30 volts  (just an example) at all times to the charge controller, you would then fire a relay when it gets to say 60 volts (just an example) and figure out your resistance load needed and apply that to bring that 60 volts back down to 30 volts. After that relay has been fired, say the wind gets even stronger and again the voltage goes up to 60. Then you fire relay #2 and bring it back down again. As the wind lets up, the voltage starts to fall, you reverse the way it works and relay 2 drops off, then relay 1. Doing it with electronics gives you much more fine tuning capabilities. That would make it far more efficient however. It would act as a variac instead of a relay. Which if one was so inclined, they could use a variac and a small servo motor to accomplish this. Just trigger the servo motor from monitoring the dc voltage right after the rectifier.
But again, it would be easier to just find someone that likes to build computer stuff and build a controller the right way from the start. I think up this $#|+, I can't do it all by myself. LOL
Because like I said controllers available do not even have a contact for a transfer switch that keeps things on grid power until the bank is full again.
SO there are tons of improvements to be made in the controller aspect. Some of which are going to squeeze every bit of energy made out of the system. That affects the payback time of the system and helps drive sales when that number is better. It also acts as a breaking system to incorporate a "variac" type dump load. On here we talk a lot about the performance of the mills we build, but the performance of what we do with it after we make it is important too and the market seems to be lacking for those products.

[Astro]

You have a pretty good handle on what you are up against.  To me, the simplest and best is still to tie the mill directly to the battery with a rectifier between. Then the dump controller just removes excess power from the batteries to keep them happy.  A battery will even out a lot of surges from quick wind gusts. You can think of this as no wind turbine controller, just a charge controller that manages the battery. You can also direct connect solar, hydro, dc generator, wind etc and the one controller will keep the battery happy.   

 I like to have the rectifier in the house by the battery.  Then you can access the 3 phase side.  I have a secondary dump load on mine that has 3 elements in delta configuration that is triggered by high voltage on the turbine.  It acts as a power limiter in that it burns off the extra power before it gets to the rectifier or anything beyond. It heats water.  It also adds redundancy.  I used a 3 phase solid state relay, but mechanical would work fine. 

A variable dump load with different resistances would be great, but it gets a little complicated and to me isn't worth it.  The very high wind days we have so much power that wasting a bit from the turbine being clamped to a lower voltage is fine. 

These are great things to have straight in your mind before you start, but it will all be clearer when you have the turbine up running and can see exactly how it acts.  They all seem to have slightly different behavior.

Yup you know exactly what I am talking about.

[Adriaan Kragten]

There are certain optimal places for the short-circuit switch, the rectifier and the dump load of the battery charge controller.

I assume that the PM-generator is strong enough and that the safety system which limits the rotational speed, the power and the thrust is adjusted at a sufficient low wind speed. In this case the rotor can be stopped by making short-circuit in the generator winding at any wind speed. Short-circuit should be made before the rectifier to prevent damaging of the rectifier diodes because of the high short circuit-current and to prevent a voltage drop over the diodes. So the short-circuit switch should short-circuit the three phase wires and if possible also the star point of the winding as this gives the highest peak torque. The short-circuit switch should be mounted at the tower foot at a height of about 1.5 m above the ground such that it is easily accessible without the need to climb the tower. The 3-phase rectifier is mounted just below the short-circuit switch in a water tight box. So a DC current is flowing through the cable from the rectifier to the house.

The size of the cable depends on the current and the length of the cable. I have used ground cable with massive copper wires of 4 * 2.5 mm^2 for my VIRYA-4.2 which is charging a 48 V battery and which has a maximum power of 1100 W. Two cables are connected in parallel so in reality I have two times 5 mm^2 for a maximum current of about 20 A and a distance in between the tower and the batteries of about 40 m. DC current is flowing all the time and AC current is flowing only during 2/3 of the time for star rectification (see my public report KD 340). So transport of DC current is more efficient. For the same losses you need 50 % more copper for AC current than for DC current if the current is finally rectified.

The battery charge controller should be mounted as close as possible to the battery to make that the voltage drop over the cable in between the voltage controller and the battery is minimal. The real charging voltage is then almost equal to the voltage of the voltage controller and independent of the current. The dump load should be mounted against a wall at a cool place inside the house such that it can never be light up by sunlight. The batteries can be placed below the dump load.

[bigrockcandymountain]

Because like I said controllers available do not even have a contact for a transfer switch that keeps things on grid power until the bank is full again.
SO there are tons of improvements to be made in the controller aspect. Some of which are going to squeeze every bit of energy made out of the system. That affects the payback time of the system and helps drive sales when that number is better. It also acts as a breaking system to incorporate a "variac" type dump load. On here we talk a lot about the performance of the mills we build, but the performance of what we do with it after we make it is important too and the market seems to be lacking for those products.

I don't work for Midnite solar but i like their charge controllers.  The midnite classic has 2 aux outputs that you can configure to do almost anything, including a grid transfer based on state of charge, battery voltage, etc.  It can be dry contact, 12v output or 0-12v pwm output to vary a solid state relay.

Mine modulates a dump load solid state relay to use excess solar/wind watts to heat water. 

They are a $900 controller.

The outback fm60 has 1 aux output with most of the same options.  They are cheaper but not cheap. 

What i like about these is that they mppt a solar array, and still control the dump load for the turbine, all with one controller.

[Astro]

Because like I said controllers available do not even have a contact for a transfer switch that keeps things on grid power until the bank is full again.
SO there are tons of improvements to be made in the controller aspect. Some of which are going to squeeze every bit of energy made out of the system. That affects the payback time of the system and helps drive sales when that number is better. It also acts as a breaking system to incorporate a "variac" type dump load. On here we talk a lot about the performance of the mills we build, but the performance of what we do with it after we make it is important too and the market seems to be lacking for those products.

I don't work for Midnite solar but i like their charge controllers.  The midnite classic has 2 aux outputs that you can configure to do almost anything, including a grid transfer based on state of charge, battery voltage, etc.  It can be dry contact, 12v output or 0-12v pwm output to vary a solid state relay.

Mine modulates a dump load solid state relay to use excess solar/wind watts to heat water. 

They are a $900 controller.

The outback fm60 has 1 aux output with most of the same options.  They are cheaper but not cheap. 

What i like about these is that they mppt a solar array, and still control the dump load for the turbine, all with one controller.

 I have looked at those. they seem to be fairly well thought out. Maybe they should send me a controller for free and I will consult with them on how to make it better. Think they will buy that idea? LOL
My whole thing is as Adriaan said above, the amount of dump is dependent on how much it needs for the battery. It switches over to dump load gradually. All I am saying is if you do the same thing and control the incoming dc voltage to the controller and dump the excess ac voltage, you would be better off. AND since it is pretty much already transferring from charge to dump gradually, the circuit would be very similar, just instead of reading volts between the controller and battery, you would be reading volts pre rectifier and adjusting the ac volts pre rectifier to dump instead of the dc volts. If it does it gradually it would act just as a soft breaking system. With relays it would be more larger increments, unless you had 20 different relays and 20 different small resistance dump loads. Even then if you tried to dial it in tight, you would have relays constantly switching. So you would have to have your "steps" larger then you would if it were done electronically.

[Adriaan Kragten]

If the battery is almost full, only a very small part of the energy produced by the wind turbine goes into the battery and almost all energy goes into the dump load. So the dump load must be that big that it can dissipate the maximum power which the wind turbine can produce. The 200 W battery charge controller in my report can even be used without a battery. If 200 W isn't enough, one can use several 200 W dump loads in parallel using the same voltage controller. This is possible because the transistor on the cooling plate is of the type Darlington and such a transistor has a very low base current

The 200 W dump load module has an aluminium cooling plate of 3 * 250 * 500 mm. In the center of the plate there is a 300 W Darlington transistor mounted on an extra 5 mm cooling plate. In the center of each half plate there is a 100 W resistor (see photo of the original drawing at the end of the report). The distribution of the power in between the transistor and the resistors is given in figure 4 of the manual. The distribution depends on the current I. For low currents almost all energy is dissipated in the transistor but as the current increases, the resistors take over. The power in the transistor is maximal for I = 4.6 A and the dissipated power in the transistor is then 63.48 W (see table 1). So this is much lower than the nominal power of 300 W and the transistor will therefore have a long lifetime. For the maximum power of 200 W belonging to a current of about 7.3 A, about 40 W is dissipated in the transistor and about 80 W is dissipated in each resistor. So the resistors are also loaded below the nominal power.

[Astro]

"If 200 W isn't enough, one can use several 200 W dump loads in parallel using the same voltage controller."

 This is exactly what I am talking about, except say 50 watt dump load increments AND that the voltage is taken pre rectifier (or straight from the mill).
You could still have a 1200 watt heater as a dump load, but it is fed from the ac power from the mill. Then it may get 0 watts, it may get 50, it may get 500, it just depends on what the controller needs to charge the batteries.
 This also keeps the voltage after the rectifier and to the controller much more stable. It should make more heat, if that is what you are doing with the dump load, by taking the higher voltage from the mill.
 Maybe the cost to benefit ratio makes it so this idea is not cost effective?? IDK, But I do know that the breaking system on small set ups is for crap. Something needs to be figured out. I also know that if over the course of 5 years a person was able to say that they produced X amount of watts more because the mill was hardly ever in stall mode, it makes the idea of investing in a system much more attractive. I mean look, if a controller came on the market and said they took x brand of controller and their brand of controller and put them on identical mills side by side and their brand had 500kwh more of useable energy per month, guess who is winning???
One thing about the big mills is they are always spinning at the same speed. Does not matter if the wind is 45 mph or 15 mph. That is what wins the race, steady speed.
It just seems like a small 6 point plc could accomplish this idea and so the cost to do it can not be all that much more.

[DamonHD]

How full the battery is is not just measured by current voltage: previous recent history over hours or days may be important.  Also  internal impedance, temperature, etc.

Efficiently using power of wild AC over a wide range of voltages is not trivial because of V^2 / R amongst other things...

Rgds

Damon

[Mary B]

"Fully charged" may be harder to define then you think, and you may also waste a lot of energy that the battery cannot absorb as it gets nearly full.

Rgds

Damon

I use one of the Morningstar relay drivers to control things https://www.morningstarcorp.com/products/relay-driver/ they are programmable for what each port can do(to a point, go read the manual!) Works well to turn off inverters based on battery state of charge.
In a relay system yes because it is absolute. But in a computer controlled or a electronic system, no. Because as it starts needing 3 amps or 10 amps for charging, it would make up that power by dropping off the relays to the resistance loads. You might be charging and have one or two resistance loads going, you might be charging hard and have no resistance loads going or you might not be charging at all or very little and have several resistance loads going.
With relays you would have to monitor the incoming voltage after the rectifier and before the charge controller and vary your resistance loads by that number. So if you wanted say 30 volts  (just an example) at all times to the charge controller, you would then fire a relay when it gets to say 60 volts (just an example) and figure out your resistance load needed and apply that to bring that 60 volts back down to 30 volts. After that relay has been fired, say the wind gets even stronger and again the voltage goes up to 60. Then you fire relay #2 and bring it back down again. As the wind lets up, the voltage starts to fall, you reverse the way it works and relay 2 drops off, then relay 1. Doing it with electronics gives you much more fine tuning capabilities. That would make it far more efficient however. It would act as a variac instead of a relay. Which if one was so inclined, they could use a variac and a small servo motor to accomplish this. Just trigger the servo motor from monitoring the dc voltage right after the rectifier.
But again, it would be easier to just find someone that likes to build computer stuff and build a controller the right way from the start. I think up this $#|+, I can't do it all by myself. LOL
Because like I said controllers available do not even have a contact for a transfer switch that keeps things on grid power until the bank is full again.
SO there are tons of improvements to be made in the controller aspect. Some of which are going to squeeze every bit of energy made out of the system. That affects the payback time of the system and helps drive sales when that number is better. It also acts as a breaking system to incorporate a "variac" type dump load. On here we talk a lot about the performance of the mills we build, but the performance of what we do with it after we make it is important too and the market seems to be lacking for those products.

[Adriaan Kragten]

What I know about lead acid batteries is the following. Lets take a 12 V lead acid battery as example. For a higher nominal voltage, all voltages have to be multiplied with the same factor.

The open voltage of a 10 % full battery is about 12 V. The open voltage of a 90 % full battery is about 12.6 V. The open voltage increases strongly above the 90 % charging state and above 90 % charging state you get strong water separation in hydrogen and oxygen. The maximum charging voltage which can be maintained for a long time without strong water separation is 13.8 V. So if you charge a 10 % full battery with a voltage of 13.8 V, the voltage difference is 1.8 V. If you charge a 90 % full battery with a voltage of 13.8 V the voltage difference is 1.2 V. This is the reason why the current decreases if the charging state of the battery increases and if the charging voltage is maintained at 13.8 V.

A problem with the open voltage is that it can only be measured correctly after at least an hour of no charging or no discharging. So one can't find the charging state of the battery during the charging process by shortly disconnect the battery from the charger. The only accurate way to measure the charging state is to measure the density of the sulphuric acid.

So the current which is possible at a charging voltage of 13.8 V depends on the charging state but it also depends strongly on the battery capacity. A 200 Ah, 12 V battery will have a ten times higher current as a 20 Ah, 12 V battery at a charging voltage of 13.8 V if the batteries are of the same type and if they have the same charging state. So to prevent that a large part of the energy which is produced by the wind turbine is flowing into the dump load, one should use batteries of a large capacity and prevent that they reach the 90 % charging state. This can be done by using the energy for useful purposes or by changing batteries if one battery is 90 % full.

I have an Optimate 3 charger for the 12 V gel battery of my BMW R80RT motorbike which uses the principles as described above. I don't use the motorbike often so every first day of the month I charge the battery for one day. The battery isn't empty when I start charging and it starts with a low current showing a green light. The charging voltage increases slowly and after about one hour, it reaches the 13.8 V level. Next the battery is disconnected automatically and a green light is flashing. The open voltage slowly decreases and after may be an hour or longer it reaches the 12.6 V level. Next charging starts again until the 13.8 V level is reached. This way of charging gives a longer lifetime of the battery and less water separation than a constant charging voltage of 13.8 V. The fact that I charge only one day per month also gives a longer lifetime than connecting the battery permanently to the charger because a battery is wearing during every hour that it is charged.

[Astro]

Just wanted to let you all know I am reading your comments. I got a little busy in the shop with some things and figuring out how much insulation to blow in the attic of the shop.
I have an old electric radiator fan that was making some noise, so I tore it apart to have a look. It works good now. I am going to get a radiator and mount that electric 12v fan behind it and then use a small 12v pump to pump the 55 degree water from my cistern through the radiator and wa la air conditioning for the shop. I have not decided if I want to use a coil of pex pipe and glycol water mix through the pipe or just pump the straight water. Right now for cost wise, I am thinking just straight water from the cistern, then just drain the lines in the winter. Then my plan is to get a shallow well pump to also pump out of the cistern to water my garden/grass. But I would rather have ac in the shop, and the cistern is big enough to do both. But the ac for the shop comes before the shallow well pump.
It is only 20x20 so if it is insulated, it should not take much to help keep it cool in there.
 I have a speed controller for the fan and will have to get one for the pump as well. I have not taken any amp readings so I have no idea how long it will run on a charged battery, so that is my next step, now that I have a working fan motor to test on. Also have an old wash machine I stripped of any goodies so I can get it out to the dump. It has a small 120v pump, but I would prefer 12v. I do not know if it will pump a head of 12-15 ft either. My plan was a sprayer pump.
100 different projects and only 1 me. I am sure ya'll know how it goes.

[Astro]

Well at 3 amps (12vdc) my radiator fan is starting to move some serious air. I did not check wide open full voltage amps, because that is way to fast. With that and the nameplate amperage of the pump, I should be pulling around 8 amps of 12vdc to have air conditioning in my garage. That is pretty much an entire day with no recharging going on, off of a 100 ah battery!!!!
Now I will scrounge up some sort of thermostat, to turn it on and off.
 I have an old clothes line post that is firmly anchored (I know we pulled the other one) that is heavy duty 3in pipe out back of my garage. My plan is to cut the top off and stick a pipe down inside of it and mount a small mill on that post. I am also going to put up the mill I am making currently out in the most windy part of the yard. My plan is to have one smaller system for the garage and one bigger more dependable system for the house. Hence the ordering extra stuff, because I already know I will be building more then one mill.
 In the garage, I plan to run dc led lighting and this homemade ac unit. I might also end up running another small dc fan for the pellet stove I plan on putting in.
I have noticed as I get older that I like led lights and I need it lit up like the sun to be able to see. Lol. I even have this head band magnifier with a led flash light mounted to it. Old people were right all along writing is getting smaller.