For LEDs, is there a device that maintains a steady "power" output, interms of watts or amps?

[rogervan]

This is for small voltage/wattage LED's. Apparently LEDs hooked up in series can have power fluctuations that can break the them. My ultimate question is are there power regulators that don't change the voltage, and are they small enough to regulate a power output of around 120 mA?

My design is to use 3 LEDs at "four V max" and 30 mA each. So if I hook them up to a 12 volt source at 120 mA, it seems I won't have to add any resistors. But the site that sells them says this in-series design can cause I (capital I) to fluctuate. So do I also need a power regulator?

My project is to make a UVB source of 380 nm LEDs, with a reasonable number of LEDs (3 LEDs at 4 volts/30mA using a 12V transformer) is all I can think of right now) to make the most out of the 12V power supply without too much resistance which would waste power and, worst of all, create an unwanted heat source. I have a UVB meter that has it's strongest response to 380 nm in the UVB spectrum in terms of uW/cm^2. I will have to build one of these 3-LED sources of UVB so I can measure their output on this meter. Then I would know how many of these units to put side-by-side to create a sum total of 50 uW/cm2, which is my goal.

Any input is much appreciated.

Roger

 

[Arouse1973]

Using a constant current source will ensure the correct current is maintained for the LED string. It adjust it's output voltage to do this, so some of the LEDs may have a different voltage drop but will all draw the same current. You set the current to what you need and that's it.
Adam

 

[sundy]

Yes regulators are available for less than 120mA. A regulator with a higher rating will not output any current that is not drawn by the load. So the load determines the current not the regulator. Well too small a regulator will limit the current until it smokes.

Your idea of 3 LED's in series with no resistors is flawed. The '1st' LED will drop more voltage than the second etc.

I was under the impression that LED's need a current limiting resistor?

A Zener diode might serve your purpose.

 

[sundy]

Arouse1973's answer is better than mine.

 

[(*steve*)]

Arouse1973's answer is better than mine.

Yes, waaaay better.

I suggest both you and the OP go and take a look at the LED resource we have.

 

[rogervan]

Arouse1973's answer is better than mine.

Using a constant current source will ensure the correct current is maintained for the LED string. It adjust it's output voltage to do this, so some of the LEDs may have a different voltage drop but will all draw the same current. You set the current to what you need and that's it.
Adam

The formula I'm using is easiest to describe as using 12V input and 4Vmax/diode,, each with a 0.03A load.
It goes V= 12-4-4-4, all over I, which is 0.03A. So I get zero for the voltage, and thus the formula does not work, or it shows that these LED's use all the voltage, drawing 0.03A.

I looks to me like I am making some fundamental mistake, since the numerator is zero, with a finite denominator, This is not math. I was trying to use up all the V so no resistor would be required. Then I got the warning about fluctuations in I, which seemed to me requires a current regulator.

Can someone make a link to a current regulator such as this?

Can someone direct me to the information at Electronicspoint about LEDs?

 

[Harald Kapp]

My design is to use 3 LEDs at "four V max" and 30 mA each. So if I hook them up to a 12 volt source at 120 mA,

On top of the above answers: When you connect 4 LEDs with 30 mA each in series, the total curent is still 30 mA, not 120 mA (current in a series circuit is the same through each part of the series circuit).

 

[(*steve*)]

Here is the resource: https://www.electronicspoint.com/resources/got-a-question-about-driving-leds.5/

 

[Arouse1973]

The formula I'm using is easiest to describe as using 12V input and 4Vmax/diode,, each with a 0.03A load.
It goes V= 12-4-4-4, all over I, which is 0.03A. So I get zero for the voltage, and thus the formula does not work, or it shows that these LED's use all the voltage, drawing 0.03A.

I looks to me like I am making some fundamental mistake, since the numerator is zero, with a finite denominator, This is not math. I was trying to use up all the V so no resistor would be required. Then I got the warning about fluctuations in I, which seemed to me requires a current regulator.

Can someone make a link to a current regulator such as this?

Can someone direct me to the information at Electronicspoint about LEDs?

Here is a simple circuit with only a few components. There are better devices now for doing this, but this configuration shown bellow should work fine for you as long as you don't stress the device beyond it's max limits. The data sheet will give you the electrical characteristics of the device.
Adam

 

[Sunnysky]

You can model this UV Diode as Vf=2.8 +If*18Ω
THen ensure your Vreg is stable to adjust Series R

This is my method which has worked for decades from sources with tight tolerance. Your mileage may vary....

Thus is using 12.0V Reg and choose MOSFET switch with 0.2V drop
(11.9 -3*2.8V)/30mA=Rtotal= 117 Ω then subtract LED ESR = 4x18=72Ω
and current limiting R = 117-72 =35Ω which should drop 1.35V plus tolerance stackup.

For more accuracy using LDO with 2V drop or 14V and CC mode as above.

Also note efficacy of LED's may have a wide tolerance (2:1) on mW/sr output and your flux density depends on gap to surface area.

 

[Herschel Peeler]

The formula I'm using is easiest to describe as using 12V input and 4Vmax/diode,, each with a 0.03A load.
It goes V= 12-4-4-4, all over I, which is 0.03A. So I get zero for the voltage, and thus the formula does not work, or it shows that these LED's use all the voltage, drawing 0.03A.

I looks to me like I am making some fundamental mistake, since the numerator is zero, with a finite denominator, This is not math. I was trying to use up all the V so no resistor would be required. Then I got the warning about fluctuations in I, which seemed to me requires a current regulator.

Can someone make a link to a current regulator such as this?

Can someone direct me to the information at Electronicspoint about LEDs?

That "4 V maximum" is maximum, not typical. What do they actually drop at 30 mA?
Do you have a data sheet for these LEDs? Last time I priced 380 nm LEDs the price was outrageous and intensity was not so great..

 

[Sunnysky]

Read #10 until you understand