I am building a control circuit for a battery charger, which I guess is kind of straight forward if you know exactly which battery is going to be used in the charger, but I don't. The battery voltage is going to be 12 V, but the battery capacity (Ah) can vary... The charger is supplied by a windmill, whos speed varies, and the current applied to the circuit will vary.
When the voltage over the battery reaches a certain level a zener diode will make a transistor conduct, and the current will be "shunted", so that the battery don't overcharges. (After that it will trickle charge)

Don't I have to limit the current applied to the circuit, so that the charging speed don't get to fast? How do I do that when the battery capasity can be between like 5 Ah and 30 Ah? I have been told lately that I don't have to worry about that... but I don't understand why?

 

It's very bad to charge a battery too fast. This raises the internal temperature of the battery, which greatly increases chemical activity, which shortens the life of the battery.

If you don't know the AH capacity of the battery connected to the charger, you'll have to measure it using a uC's ADC; voltage vs current over time during a discharge cycle, and remember the results somehow.

Then charge the battery at a recommended charging rate.

Note that the battery internal temperature should be considered when charging to determine the actual state of charge. A battery at lower internal temperature is charged to a higher voltage than when it is at a higher temperature.

 

Speaking of charging rates...

There are several phases to charging a battery:
1) The "bulk" charge phase - this is where the battery is >15% discharged, and you are charging it at about 1/8 the AH rating or less.
2) The absorption charge phase - charging at a reduced rate to "top off" the batterys' charge.
3) "Float charge" phase - you maintain a float voltage over the fully charged voltage. The float voltage will vary with the battery chemistry and internal temperature.
4) Maintenance charge - occasionally, an overcharge condition may need to be applied for an hour or so, depending upon battery chemistry. This is to alleviate sulphation and to ensure the electrolyte is well mixed. Gel-cell batteries should not be subjected to such maintenance charging, as the gel will develop bubbles, permanently reducing battery capacity.

 

My problem is that the charging circuit and the windmill is going to be put out somewhere, and pepole is going to use it to charge their 12 V Deep cycle batteries.. I have no control control over which battery capacities they will be trying to charge.. (And they will for sure try to charge whatever batteries they have) which makes it really much more difficult to charge the battery at the proper speed.. since the speed recommended for each battery type is different.

Then I was told not to worry about that, because something about the windmill power curve and speed or something... (I really didn't understand it.. as you can see..), and I can't for the life of me figure out why a single value fuse can protect the battery from being over charged, when the charging current limit depends on the battery Ah...

 

You are making it harder than it really is. Imagine charging a lead-acid battery (SLA or flooded-cell) from a lab power supply; the kind that has a "voltage adjust" knob, and also has a "current limit" knob.

The correct way to do it is to set the voltage knob (with the supply not connected to anything) to the recommended "charging voltage", corrected for ambient temperature for the specific battery. Here is a minor problem, because at room temperature (70degF), the charging voltage for a flooded-cell lead-acid 12V battery is 13.8 to 14.2V, while for a SLA, it is 14.4 to 14.7V (see battery manufacture's web site).

You also need to set the "current limit" knob (with the supply short circuited) to the maximum allowed charge rate for the specific battery. A large (50 to 75AH) flooded cell will accept charge rates of up to 50A, so likely your lab supply (or windmill) wont even come close to that, so set the knob for maximum... For a small 3 to 10AH SLA, the maximum charge rate is limited by the manufacturer to 2 to 4 A; more for larger SLAs, so set the "current limit knob" to match the battery.

Now if you connect a (mostly discharged) lead-acid battery, you will see the battery terminal voltage start initially low, like 12V, i.e. it "pulls down" the power supply's output voltage. The current is initially limited electronically by the current limiter in the power supply to whatever you set it to. As the battery accumulates charge the battery voltage slowly climbs (over several hours) until it reaches the voltage that was set above. At that point, the voltage stops climbing, because it is now being limited by the power supply's voltage regulation circuit, however, the current slowly begins falling toward zero (takes a few more hours).

This is the normal charging regime for lead-acid chemistry batteries. Your windmill-powered charging regulator has to do exactly what the lab supply does in the example above. It needs to have a "voltage regulator", and it needs to have a "current limiter" to protect small batteries from excessive charge rates.

The biggest problem you have is that it needs to have an independently switch-selectable charging voltage (Flooded or SLA), and it needs to have a switch-selectable charging current limit (2, 3, 4,...10, or unlimited).

If you need to make this "idiot proof", then limit the voltage to 14.2V (that way it will charge flooded batteries well and SLA's poorly, but wont damage them), and limit the max charging current to 2A ( that will not cook a small SLA, but it will take forever to charge a large flooded-cell battery).

 

Thank you so much for the switch-system idea, I found a way to apply that to my circuit so that each different battery capacity that is going to be charged can get the appropriate current.
My next issue is to decide at what current to shut down the windmill. To figure that out I need to decide at which speed to charge the battery. So how fast do one want a 70-100 Ah battery to charge? Is 7-8 hours too long?

 

Its NOT a matter of HOW LONG; it is a matter of watching the battery terminal voltage, and the battery charging current. As I said in my long post above, when the battery reaches the set charging voltage, the charging current begins dropping. The battery is fully charged when the charging current drops below ~10% of the initial current that the battery was taking while the voltage was initially climbing. This could take minutes, or it can take days, depending on the battery capacity, the current limit, and the initial state of charge.

Read this.

 

If you are dealing ONLY with lead acid batteries, you can use a ΔV/ΔT measurement, or a temperature probe that is physically connected to the battery. If the voltage stops rising for a 2 minute period, or the temperature goes up a few degrees in a minute, then the battery is full.

The same concepts apply for charging NiMH, NiCd, and Li-Ion, though the parameters and requirements for accuracy are much more strict to avoid destroying the cells.

 

If you are dealing ONLY with lead acid batteries, you can use a ΔV/ΔT measurement, or a temperature probe that is physically connected to the battery. If the voltage stops rising for a 2 minute period, or the temperature goes up a few degrees in a minute, then the battery is full.
...

Caution! the ΔV/Δt method only works if you are using a "contant-current" charger; it does not work with a "voltage-limited" charger, and is not a very reliable method of terminating battery charging. This charging algorithm is used in most commercial battery chargers.