resistorless LED

[obi-one]

First of all... yeah, I know, up there I found and read carefully about the topic, but since the last post posted there was complaining about someone elses resurrecting such an old thread, I decided to start a new one. I apologize for it. =P

The thing is, I was checking the wiring guide and I find it somewhat erratic... maybe... I don't know, just an idea.

As you probably all know, a LED acts, when forward biased, pretty much as a closed circuit. Near to zero resistance. Or at least that applies for regular LEDs.

Now, the Wiring guide says that for a Luxeon V led, ranked at 6.8V 700mA, a 1 Ohm resistor should be in series in the LED with a 7.2 voltage source...

If I recall correctly, Ohm's simplified law says: V/R = I. So, supposing a 7.2v unlimited current source, I'll be feeding the LED with 7.2 Amperes!!!!!!!!!!!!! ten times the nominal current.

So, I'm assuming either two things of this:

1. Luxeon LEDs DO NOT act like regular leds, i. e. they DO have an internal considerable resistance. And doing the math, I'm thinking of a 10 ohms resitance. Dynamic resistance in function of temperature for sure, but nevertheless always close to 10 ohms.

or

2. The battery pack of 7.2v referred in the wiring guide can output a maximum of 700mA at any given time. Which would be a weird thing, not to mention that would make the buckpacks and drivers almost completly uneccesary. So I'm going for my first guess.

So if luxeon leds already have their inner resistance, the resistors at the wiring guide aren't really that necessary.

Because anyway, 1 ohm or 3.3 ohm (signaled for 9v) are ridiculous values of resistors. Chances are high that a poorly done soldering from my part ends up giving me those couple of ohms.

So, I wish somebody can tell me about my deductions are right or completly wrong.

[psab keel]

Using a resistor or driver with an LED is ideal. The thing is that because batteries can be higher or lower than their nominal voltage when at full charge, and LED's (especially Luxeon LED's) are binned specifically for forward voltage, you can either over or under drive your set up depending on the resistor used.

I personally no longer use resistors to drive my Luxeon's as a driver regulates the current much better.

[neophyl]

You arent taking into account voltage drop across the led. Even 'normal' leds have a voltage drop, thats why you see the little keyring led lights that just have battery/led with no resistor and work fine too.

[FenderBender]

I have yet to do it with a Lux V, but we've made damn near close to 200 PVC stunts and havent used resistors on any of them. I don't believe that Ultra does either. We do however, match the voltage of the LED to a proper battery pack.

[Kant Lavar]

I've not done it with a Lux V, but...

Logically, as long as the voltage you give it doesn't burn out the LED, the only problem with giving a Luxeon LED more than its normal specified amperage (in the case of a Lux V, 700 mA) is that you'll have a drop in its usable lifetime. But, as Novastar mentioned once (can't find the post now), even giving the Lux V 1000 mA won't kill it immediately - it'll just drop the lifetime of the LED by 25%.

Here's the trick, though. Novastar said that a Luxeon V, run at the recommended 700 mA, will run for something on the order of 100,000 hours. So cut a quarter off that, and you've got 75,000 hours of expected runtime, which works out to...

75,000/24=3,125
3,125/365.25=8.55578371

Just over eight and a half years of CONSTANT use. But nobody uses a saber constantly - if nothing else, the batteries have to be recharged or replaced. So let's figure... hell, let's be realistic and say a yearly average is about three hours.

75,000/3=25,000
25,000/365.25=68.4462697

So, at an average of three hours of lightsaber use per day, every day (which probably is still a bit high for the average user!) you still have the next best thing to 70 years worth of runtime on an overdriven Lux V.

Again, this isn't my original research. Novastar was the first one that I saw that pointed this out. But since I got it working, I've been driving my Lux V on an UltraSound 2.5 card at the Lux III setting (that is, 1000 mA), and I haven't seen a problem.

[neophyl]

Its not too much voltage or too much current that directly kills leds. Its HEAT.
Current passing through the led causes light and as a side effect heat, which is why we have to use heatsinks.
The more current that passes through the more heat is generated. As theres a relationship between voltage and current increasing one can increase the other too.

If your heatsink solution can handle more then you can correspondingly run higher currents through the led without as much problem.

ALWAYS have your led connected to its heatsink whenever you are testing your wiring. Its also a good idea to have it mounted when soldering your wires on too, especially if you arent used to soldering as applying too much heat for too long can also damage the led that way before you even apply power.

While the higher current less lifetime rule of thumb is generally true and can be used there is also a cut off point, where the current/voltage is just getting things too hot and once you get there the failure time is quick and sudden.

[Novastar]

I always say it, but... I love when Neophyl posts...

I learned about this idea of "going resistorless" basically from reading many past posts back in the day.

If the "theoretical" is not enough for you, consider the following videos:

LUX V -- http://www.youtube.com/watch?v=Os7TxBQfEZs
LUX III -- http://www.youtube.com/watch?v=sn9qlwvtIGk

Now it's true than on the Lux V setup--I *DID* use a resistor... but it was only a 0.75ohm resistor! Next to nothing... and it was simply because I found that a fully charged 7.4v pack allowed the Lux V to get a TAD too hot... I could feel TONS of heat through the hilt.

With the 0.75ohm + something like 7-watt capable resistor... it's perfect. The LED draws about 1280mA at full charge... and things get pretty TOASTY... but not out-of-control. That saber has been running for about a year with NO problems... and it's the brightest thing I own, hands down, no questions... it owns every saber I've ever had in brightness (including the v1 blue H-blade).

As to the Lux IIIs... well... I've had those built for longer than the Lux V one, and they have *NO RESISTORS*. As a matter of fact, royal blue is a GREAT color to go with a 3.7v cell directly, as this color tends to require more voltage than colors like Green or Teal.

The Lux IIIs are very bright, and certainly start off brighter than most all of my other Lux IIIs. As the battery capacity decreases, we see a bit more "normal" light output, and eventually... the cell needs recharging.

If I leave the Lux III sabers on for WAY too long... they do get overly hot, which is bad. But they aren't meant for marathon runs being that they are directly driven with no fancy-pants sound/light driving. Duh.

[obi-one]

The Lux V probably gets damn too hot because of when the batteries are fresh, they supply considerable more voltage to the LED that required (7.4 vs 6..

The Lux III works fine directly with the battery cause the batery, even just charged up, will deliver no more that the exact amount of voltage necesary (3.7 vs 3..

So when you put a resistor to the V, the resistor is probably holding back all the unnecesary voltage, and that's why it works a lot better with a resistor when you can not match the exact voltage for the LED. Which makes a lot of sense, because after all, LEDs are DIODES, and diodes act just like that. Considering the voltage drop across a luxeon, it could be thinked of it like a Zener Diode.

So ok, I figuered out (theoretically*) exactly how they work!!!!!!!!

*But I still gotta put my hands on the real thing, so I can start experimenting.

But now the question is, why would a driver board or buckpack help to make batteries last longer rather than a well matched voltage source, or a higher voltage with a resistor?

The only way I can think on how they make batteries last longer is because instead of sending a constant current they send a fast burst of pulses, half low, half high, efectively saving a lot of battery and making use of the fact that a LED doesn't fades instantly.

The thing is I'm trying to build my own buckpack hehehe...
Regards

[KuroChou]

I'm running a 1995 hasbro board off of 4 AAA's, and it's giving me an output of 3.9V. Is that in an acceptable range of overdrive for a white P4?

[Count Malik]

yes I would say thats exceptable, because the p4 white has a fowared voltage of 3.25 and if your battery voltage is a 4 aaa pack w/ hasbro eco. board, then the p4 is being driven good enough.