Direct drive question

[theepdinker]

First off I'm a mechanic type, not an EE.
I'm not stupid, just not my thing.
Please keep the electrical explanations simple, Thanks.

I've read most of the posts (arguments) about DD yea/nay.
I understand I can/I should'nt. Doesn't mean I won't.
The discussions argue for regulation to prevent thermal runaway.

I'm doing alot of battery research at the moment & internal resistance is a big issue.

Here's my question.
Since LEDs use resistors to control current, why not use existing high internal resistance cells to our advantage.
Does it require that much more resistance?
Are we talking fine lines, or can these cells be used to provide a reasonable buffer zone?

Theepdinker

 

[KevinL]

As long as you have a DMM, understand what you are doing and measure often and in the right places you should be ok. I've sized a few resistors for my lights, and most of the time I end up undershooting my target current because my calculations are conservative and do not take into account battery resistance. That's ok with me though, 1050mA on a Lux3 is still plenty of current being thrown around.

Take for example the famous DD Mag3D with Lux3 - the Mag has around 0.4 to 0.5 ohm of resistance internally thanks to the switch, spring and inner works, Duracells have around 0.13 ohm x 3 cells, that gives you a total of close to 0.8 ohms and 1125mA of current through the Lux (Vf = 3.6). The Lux has its own resistance too and lower Vf Luxeons will pull more current.

NiMH cells on the other hand may cook the Lux very efficiently due to their very low internal resistance..

So if you understand what you are doing there's no reason why not to DD. It's a design choice, really. That's the fun of modding - you get to design, build, and validate.....or mess up and destroy a light like I just did /ubbthreads/images/graemlins/tongue.gif

 

[liquidium]

I'm still learning, but I'll take a stab @ this.
The rub is that most decent battery chemistries possess low to very low internal resistance: NiCad, Lithium Primaries, Lithium rechargeables, NiMh.
I think you can exceed the LED's design Vf (forward voltage, that is - the starting voltage the emitter was designed around) with a simple resistor without the need for a driver/ regulator in most cases, but I'm not good at calculating resistor values.
Tell someone here about the LED and battery configuration you'd like to use and you'll probably get an answer from someone who knows more about it.
Hope this helps. /ubbthreads/images/graemlins/icon6.gif

 

[illumiGeek]

If you have good heat sinking over-driving a Lux3 will not have a dramatic affect on its functionality.

A 3-cell DD on a Lux3T with a K voltage bin will usually result in a current of around 1.5A at the emitter. This is about 50% over the 1A spec max, but with good heat sinking the temperatures seem to be well within spec.

I'm sure someone could demonstrate that over the long run there is some negative affect on LED life, but when you consider how long these things are supposed to last you are still talking about a light that will last a very long time.

If you are really overly paranoid about such things you could use an L or M voltage bin which would run closer to spec.

I've built several DD 3-cell Mag Lux 3W mods and I can tell you that they are a relatively easy and very rewarding mod. Freakishly bright, durable, reliable and long running.

I mean, I like a nice regulated light as much as anyone. I own a VIP and love it. But I find the simplicity of the Mag DD mod equally appealing. For about $50 (give or take $10) anyone with minimal skills and limited tools can build an LED light that will kick the crap out of most fancy regulated lights that cost loads more.

 

[KevinL]

That's why I love the Mag. I'm very leery of other lights, especially plastic lights, because they simply do not provide adequate heatsinking. Scalding hot heatsinks after 5 minutes of runtime = bad for the Lux, bad for you. Mag conversions also give you a huge variety of choices - Luxeon 1W (but WHY? unless it's a red), Lux3, Lux5, two choices of professionally made heatsinks, quite a few choices of lens, and yeah, DD, resistored, or regulated.

You can start with a DD, then move on to resistored or regulated as your needs dictate. My lights are built this way, if I ever need to add a regulator, it's a simple matter of just shelling out the money and waiting for the parts to show.

 

[theepdinker]

double clicked sorry

 

[theepdinker]

Alot of replies addressing LuxIII.
How about the 5 watt model. W bin

Theepdinker

 

[KevinL]

The whole idea is to hold your current within range. It's all about the current. I would be leery of pushing a Lux5 too hard because of its low lifespan, maybe 1000 hours. Pushing it may result in degradation - I wouldn't know but it wouldn't be a good sign to blow my first Lux5.

My planned Lux5 setup is a 3D Mag with 3 x 2AA battery holders running NiMH, the problem with this is that NiMHs off the charger have a significant voltage spike, multiplied by six of them makes the voltage fluctuate far more than 3/4 cells driving a Lux3. On top of that I think the Lux5 is not so forgiving to higher currents too. I personally would resistor it just to be safe because it is so darn expensive and W bins are so hard to come by. But even with a resistor, the current still swings more than I'd like it to. Maybe time to regulate... /ubbthreads/images/graemlins/smile.gif

But like I said before it all depends on your individual needs and personal preferences (I'd prefer a tighter spread of current). Some of CPF's nicest mods are direct drive - many are quite happy driving the Lux5 at 1.5 amps (more than 50% overspec) in Hotbeam's awesome MR-X. I believe the MR Bulk SN2s are also direct driven Lux5s.

 

[RussH]

I wouldn't try to direct drive a Lux5 at much over it's rated current (750ma?). For one thing, the current could increase quite a bit when it heats up. Then it will heat up more, and the current will increase quite a bit more. Then it will heat up more....thermal run-away isn't certain, but the probability increases drastically with over-current. This concerns me more than the 1.43v fresh off the charger NiMH voltage. This high voltage would drop fast in use. 6 x AA may be a little high for DD even at 1.2v for most Vfs. I think I'd size my resistor based on 1.35v, or 1.0 ohms @ .75 amps (8.1-6.8=1.3, 1.3x.75=.97). A 5w, 1 ohm resistor is getting a little large.
However, with the hotlips heatsink and magic resister, the MR-X (it's not direct drive) seems to deliver something like normal life with 1.5 amps. And last I heard, Hotbeam still had his going strong at 2amps. I've no idea how many hours he has on his. Mine has about 40 hours. Some of the people who have these are pretty far up there in hours. Maybe it's time for a follow up report.

 

[evan9162]

The problem with any non-regulation circuit is that we are now more often seeing a shift in Vf of luxeons over the lifetime. I've said this a few times before (souding like a broken record yet? /ubbthreads/images/graemlins/tongue.gif ) but I have a Lux5 which the Vf dropped by about 0.5V in the span of a few months of light intermittent use. Even a resistored light (which this was) set for a specific current will see a current increase if the Vf of your LED drops. A DD light will see a more drastic increase in current due to a Vf drop.

 

[cgpeanut]

KevinL1,

Very informative /ubbthreads/images/graemlins/thumbsup.gif I'm in the process of modding my first 4d-dd mag thanks to you for posting that awesome cookbook /ubbthreads/images/graemlins/bowdown.gif /ubbthreads/images/graemlins/bowdown.gif /ubbthreads/images/graemlins/bowdown.gif
I got most of the parts except for the 3 watt resistor, Is it ok to use a 10 watt 1 ohm resistor instead? I could not find any 3 watt resistors Rat Shack only have the 10 watt in stock also, measuring the current draw of the 4d mag what value sense resistor do you recommend will a 1/4 watt resistor suffice?

Thanks in advance