LED lighting on a budget!

[BluScrnOdeth]

maybe a little. There is enough airflow in there that its not a concern to me. Plus the sides have air ports on the side where the coolest air enters so its getting a fresh dose of cooler air before it can be warmed up by the heatsink. The heatsink barely gets warm too. I was expecting it to be warm, like warm water you wash your hands with, but nope, its staying pretty cool and i touched an LED to make sure it had good thermal conduction, the LED was warm, but not painful warm.

 

[BluScrnOdeth]

Ok, i found my problem. One of my RB LEDs got in the mix that i damaged a while back. Thats what was smoking. I got a meter to them today and they are running at 3.2V the CW's are running at 3.22V so all is good. and that is with one 2.5ohm 5watt resistor on each array.

 

[kcress]

The voltage across the LEDs is of little importance. It's the current through them that can toast them. Have you measured it? You can measure it across your resistors if you know their exact value.

Vmeasured_across_resistor / 2.5 Ohms = current running thru the string.

 

[reefmusic]

The voltage across the LEDs is of little importance. It's the current through them that can toast them.

This is true. Your driving circuit needs to control/limit the current. The LEDs will have a forward voltage drop that is fairly constant and varies only a little as the current changes.

 

[TheFishMan65]

True but at 3.2 volts unless the LED is way off the graph he is a fine operating range.

 

[BluScrnOdeth]

The voltage across the LEDs is of little importance. It's the current through them that can toast them. Have you measured it? You can measure it across your resistors if you know their exact value.

Vmeasured_across_resistor / 2.5 Ohms = current running thru the string.

very little importance? according to the cree datasheet, the mA's it consumes is based on the voltage supplied. How accurate the LED is compared to the graph, i dont know. But i'll measure it once i get home tonight.

 

[TheFishMan65]

I know you can see a large variation of the voltage with the same current. I had a string and measured 3.3 to 3.6 or maybe 3.2-3.7 for the Vfs. It is back in one of these posts. IMHO voltage is not a accurate measurement for getting the current.

 

[reefmusic]

Guys, honestly, you need to control the current, not the voltage. They will drop whatever voltage their physical characteristics determine. As long as your source voltage exceeds the LED total forward voltage drop, and you control the current as per the datasheet, you'll be operating them within specifications.

 

[BluScrnOdeth]

Guys, honestly, you need to control the current, not the voltage. They will drop whatever voltage their physical characteristics determine. As long as your source voltage exceeds the LED total forward voltage drop, and you control the current as per the datasheet, you'll be operating them within specifications.

I'm not sure what you're talking about because controlling the led Vf determines its mA. Within a certain % of course.

 

[TheFishMan65]

I think what everyone is trying to get at is:

A small change in voltage can make a large/huge change in current SO you are better off measuring the current.

 

[BluScrnOdeth]

ah, yeah, of course but that means that you need to regulate the voltage, not the current the best that you can. to get rid of that slight change. right? plus i have a 1A fuse on there, so if anything happens, it will blow hopefully before the LEDs do.

 

[reefmusic]

ah, yeah, of course but that means that you need to regulate the voltage, not the current the best that you can. to get rid of that slight change. right? plus i have a 1A fuse on there, so if anything happens, it will blow hopefully before the LEDs do.

Have a look at commercial LED drivers. All the ones I see commonly in use are constant current drivers. It is much easier to regulate the current, because. as said above, a very tiny voltage change will quickly endanger the LED. Controlling the current is easy, and the LED will drop the voltage (as read from that graph) that corresponds to the current through it.

 

[reefmusic]

I know you can see a large variation of the voltage with the same current. I had a string and measured 3.3 to 3.6 or maybe 3.2-3.7 for the Vfs. It is back in one of these posts. IMHO voltage is not a accurate measurement for getting the current.

Not the LED voltage, I agree. That's why it's best to put a small resistor in series and measure the voltage across that.

 

[BluScrnOdeth]

The voltage across the LEDs is of little importance. It's the current through them that can toast them. Have you measured it? You can measure it across your resistors if you know their exact value.

Vmeasured_across_resistor / 2.5 Ohms = current running thru the string.

I measured it. 2.77V

so thats 2.77/2.2 (sorry it was a 2.2ohm) = 1.23

i'm assuming that means that its drawing 1.23A? and thats not a good thing considering my fuse is set for 1A and the RB LEDs are supposed to only handle max 1A

 

[BluScrnOdeth]

ok, well i rechecked the resistor's actual resistance and it says that the 2.2ohm is actually a 2.3
so 2.77/2.3=1.21

 

[TheFishMan65]

Can you break the string and actually measure the current as a double check. Either way I would be turning down the power supply. Or add another resistor or 2.

 

[BluScrnOdeth]

what do you mean by break the string? IF you want me to measure the mA, i cant do that because my meter only does 200mA max I'll have to go get my dads which i think can do something like 3A

 

[TheFishMan65]

Yes, open up the string and insert the meter, but your right 200ma won't work. I would get it down until you can verify - especially since I think you are probably over.

 

[BluScrnOdeth]

i agree, im going to play with the voltage adjuster for a little, try 47V and then 46V and see what happens.

 

[BluScrnOdeth]

ok, i dialed it down to 46V, it put it in the 800-900 range