I need input on how to build a voltage supply circuit for LEDs using wind power.

[dchale]

I am an electronics novice (took Control Data's micro electronics course way back in the mid 80's and never used the knowledge gained) trying to build a circuit to power LEDs using a windmill and a micro generator.

Generator Specifications:
Output voltage : 3V-24V
Output current : 0.1A-1A
Rated power : 0.5-12W
Rated speed : 300-6000 rev/min

The Led's I'm wanting to use are:

5mm LED Diodes round head and clear lens

Wavelength Range:
Qty
(2) Red: 620-630nm
(6) Blue/White: 460-470nm

I'm rather clueless on how to create a stable voltage and current to these two sizes of LEDs with a supply current that varies based upon the strength of the wind.

Also, I've been comparing the pros and cons of series verses parallel circuits and am leaning towards parallel thinking that if a bulb fries, then the rest of them would still be lit. I understand that I would need a certain sized resistor in line to each of the LEDs in a parallel circuit but that's okay with me.

Any help would certainly be greatly appreciated.

 

[Audioguru]

LEDs have different forward voltages even if they have the same part number. When you parallel LEDs then their forward voltages all must match. In parallel the one with the lowest voltage lights well and the others either do not light or light dimly and the LED with the lowest voltage takes most of the current and fails, followed by the next one then the next one. If a series LED string fries then its current-limiting resistor was calculated wrong resulting in too much current. A fried LED might blow open or become shorted. You can match the voltages of the LEDs if you buy a few thousand and measure each one. In series then you must be careful that they all do not have a high forward voltage then be dim, or they all do not have the same low forward voltage and conduct too much current then fail.

You cannot design the circuit until you know the forward voltage of each group of LEDs and the brightness (amount of current) you want. Also you must know the minimum and maximum loaded voltage and current from the generator.

 

[dchale]

LEDs have different forward voltages even if they have the same part number. When you parallel LEDs then their forward voltages all must match. In parallel the one with the lowest voltage lights well and the others either do not light or light dimly and the LED with the lowest voltage takes most of the current and fails, followed by the next one then the next one. If a series LED string fries then its current-limiting resistor was calculated wrong resulting in too much current. A fried LED might blow open or become shorted. You can match the voltages of the LEDs if you buy a few thousand and measure each one. In series then you must be careful that they all do not have a high forward voltage then be dim, or they all do not have the same low forward voltage and conduct too much current then fail.

You cannot design the circuit until you know the forward voltage of each group of LEDs and the brightness (amount of current) you want. Also you must know the minimum and maximum loaded voltage and current from the generator.

Forward Voltage: R/Y 2.0-2.2V; B/G/W 3.0-3.2V; Max. Current: 20mA.

 

[dchale]

Forward Voltage:
Red 2.0-2.2V;
Blue White 3.0-3.2V;
Max. Current: 20mA.
Wattage 0.06 watts

 

[kellys_eye]

You could make a system based on the maximum voltage output (and calculate the required series resistor using that voltage) and be sure that you'll never exceed the maximum parameters of any of the LEDs.

You can parallel different LED colours and use separate resistors in each parallel branch - each resistor will (ideally) be different values but only by small amounts - actually small enough differences to not even bother. At 24V, using a 1k resistor will limit current to around 20mA in a white/blue LED and be every-so-slightly higher in a red/green/yellow LED but not so you'd notice.

At lower voltages the LEDs will still light albeit with reduced efficiency but it's for you to decide if this is an acceptable solution. The only 'advantage' such an approach gives is that you don't need any regulation circuitry.

To get the maximum POWER from the system you need a MPPT type system which is waaaay over the top for such a simple (and low powered) circuit and would itself take away some of the generated power in its 'working'.

 

[Audioguru]

The voltage and current from the generator varies so much that the LEDs will be a little dim and bright.
If you plan to have the LEDs light fairly well only when the generator output voltage is 15V to 24V then three 2V red LEDs and a 910 ohm series resistor lights them with 19.8mA or 19.1mA when they are 2.2V and the generator produces 24V. When the generator voltage drops to 15V then the three 2V LEDs light with 9.9mA (a little dim) or 9.2mA (more dim) when the three LEDs are 2.2V.

With two 3V white or blue LEDs and a 910 ohm resistor the current is 19.8mA or 9.5mA when the LEDs are 3.2V. when the generator voltage drops to 15V then the two 3V LEDs light with 9.9mA or 9.5mA when the LEDs are 3.2V.

The 910 ohm resistors should be 0.5W which will get pretty warm but not too hot.

 

[dchale]

You could make a system based on the maximum voltage output (and calculate the required series resistor using that voltage) and be sure that you'll never exceed the maximum parameters of any of the LEDs.

You can parallel different LED colours and use separate resistors in each parallel branch - each resistor will (ideally) be different values but only by small amounts - actually small enough differences to not even bother. At 24V, using a 1k resistor will limit current to around 20mA in a white/blue LED and be every-so-slightly higher in a red/green/yellow LED but not so you'd notice.

At lower voltages the LEDs will still light albeit with reduced efficiency but it's for you to decide if this is an acceptable solution. The only 'advantage' such an approach gives is that you don't need any regulation circuitry.

To get the maximum POWER from the system you need a MPPT type system which is waaaay over the top for such a simple (and low powered) circuit and would itself take away some of the generated power in its 'working'.

Thanks a bunch Kelly... That's some great info.

 

[dchale]

The voltage and current from the generator varies so much that the LEDs will be a little dim and bright.
If you plan to have the LEDs light fairly well only when the generator output voltage is 15V to 24V then three 2V red LEDs and a 910 ohm series resistor lights them with 19.8mA or 19.1mA when they are 2.2V and the generator produces 24V. When the generator voltage drops to 15V then the three 2V LEDs light with 9.9mA (a little dim) or 9.2mA (more dim) when the three LEDs are 2.2V.

With two 3V white or blue LEDs and a 910 ohm resistor the current is 19.8mA or 9.5mA when the LEDs are 3.2V. when the generator voltage drops to 15V then the two 3V LEDs light with 9.9mA or 9.5mA when the LEDs are 3.2V.

The 910 ohm resistors should be 0.5W which will get pretty warm but not too hot.

Sweet info Audioguru… I'll be playing with these scenarios next week. This helps me a lot to know that I can build a pretty simple circuit and get by with it. Here's a pic of one model of the units this lighting will go into.


The new models in the works will have significant changes! They are all octagons and instead of the entire unit rotating in the wind, just the top roof section will rotate. I'll post some picks, (or a video) once I have perfected the system. Thanks again.