Driving LEDs in parallel.

A while ago I experimented with 10 ohm dividing resistors and two or three white leds. I had a variable max 757 voltage source. I adjusted the voltage to get about 25 mA through each led. The leds weren't exactly matched, but they were close.

As I recall, I think I got the same current distribution with or without the dividing resistors, so I said, ahh, why bother with the dividing resistors, just set the voltage so as not to overdrive the weakest led.

I think this question has been asked before but answer it again if it is redundant.

I'm trying to drive multiple Nichia White LEDs in parallel. I know that the LEDs vary in voltage for any given LED.

If I had a bank of 5 LEDs. Is there an optimum small resistor value to use in series with each LED to balance them out.

small is the key word. I want to drive it with the lowest possible voltage since this will be a battery operated device and I need maximum efficiency.

TIA.

Wayne, I think there is only one good way to use Nichia LEDs in parallel: Buy a lot of these LEDs and select them according their individual specs. Eventually you will have eg. 6 LEDs with 20mA at 3.25 volts, 10 with 20mA at 3.3 volts and so on.

Btw, your question is puzzling me ...

i've got an array of 17, some are 50 degree, some are 20 degree, some are from the paul guy, some are from an ebay purchase. unless the voltage gets very low (like 2.4~2.6V) they all seem very similar in light output. i wouldn't worry about it.

Going "by the book", driving LEDs in parallel without individual resistors is not recommended, but if they're hand-matched, you can get away with it. So just do it. Everyone else does.

Maybe I've been lucky.

An array of 12 here is made up of 4 groups of 3 LED's in series, 12v DC-DC supply. No limiting resistors in line. I don't remember the metered load (it's been a year), just that the LED's are close to 20 mA specs. All LEDs came from whiteleds.com.

Another array of 20, direct drive from 4.5v 10aH lead-acid battery, LEDs overdriven to 65 mA, all in parallel, and it performs the same.

Ditto papasan. [Edit - & Craig]

Until voltage/current drops very low, I can't see any difference in light levels. A meter might say different, but I'm happy with the results.

Maybe you should just wire 'em up and see how they look. If there's large discrepancy in brightness from one LED to another, reconsider limiting resistors to even things out.

Good luck.

Great,

Thanks all for the information.

You might want to just back off a bit on how hard you drive them. I drove 4 of them in parallel with excess heat, and had one turn "green". The rest remain fine. As long as you're not pushing them too hard, I don't think it's that much of an issue. It's more efficient to use more LEDs (but driven less) anyway.

to update...

my 17 various nichias in parallel were working very well while they were in free air. now i've put them into a fairly air-tight headlamp and i've been having problems with them. after about 30 minutes at 'full' - ~3.5V - one of them starts to get that angry aqua color that tells you it's pissed. after about 45 mintes a couple others start to complain visually.

i put a small resistor in-line on the one that overheats first but now another seems to get just as angry, although a little slower.

turning the LVR down just a little seems to work just fine though, even for 6+ hours in the waterproof housing.

i coated the whole PCB in thick coats of fingernail polish before i installed it, perhaps this is insulating the heat in too much and helping to cause these issues.

<BLOCKQUOTE><font size="1" face="Verdana, Arial">quote:</font><HR>Originally posted by papasan:
now i've put them into a fairly air-tight headlamp and i've been having problems with them. after about 30 minutes at 'full' - ~3.5V - one of them starts to get that angry aqua color that tells you it's pissed. <HR></BLOCKQUOTE>

I've got a newer model Ledda 4X, and two of its LEDs get that "pissed off color" almost the instant you turn it on; while the other two stay white. That tells me the LEDs were improperly matched; and the two that glow that electric sky blue color are probably getting 2x to 3x the current of the other two.

After the batteries have been knocked down a bit (around 30 minutes or so of intermittent use), things begin to calm down. So don't think that it's only hobbyists that can't match their LEDs - the pros can screw it up too.

I've got a newer model Ledda 4X, and two of its LEDs get that "pissed off color" almost the instant you turn it on; while the other two stay white

The original Traser Teknolite was like that- one LED went sky blue right away, (the others soon followed but that one was so bad it failed after 5 mins). The problem was so bad with just about all of them that Traser shipped new LED assemblies with current limiting resistors to everyone who`d bought a Teknolite,

FYI the Teknolites sold now all have the current limiting resistors, and are for all intents and purposaes, pretty damn decent lights. Love those little traser glow-vials in the head....