Freak LOC?

Hey everyone. I just finished putting together my LOC MXDL single-AA build (had an epiphany on how to isolate the case pin) and it's been working... fine.

Now, of course I am using incredibly low duty-cycles (because it is encased in shrink-tubing to isolate it, so the thermal transfer sucks), but should it not have died already?

The driver I am using is this one: $6.97 - 3.6V~9V 800mA Regulated IC Circuit Board for Cree and SSC LEDs (4-pack) - Flashlight Parts and Tools

I use it all the time, it's a great driver, and I modified it to output around 434mA (happened to have the resistors on hand at the time of modification to do that). This driver is a buck driver, meaning it can run off a single li-ion battery (I measured the voltage necessary to the diode to get that current, it's around 3.3V or 3.4V, if I recall correctly).

The problem arises when I tried measuring the current draw from the battery at the tailcap just a few minutes ago.... I measured it to be about 570mA! That's *way* over the safe limit of an LOC... it should have fried by now, no? And this is a buck driver, so the diode is supposedly getting even *more* current than that, because the way a buck driver works (as we all know) is by increasing current and drawing excess voltage....

Note that this test *was* conducted while using a crappy battery (3.8V or so off load).

Does anyone have an answer for this? I am very curious....

Yes, that's way over. But the ones I tested didn't die until about 650mA or so.

That just means it'll die soon. I give it 10 minutes at most.

My most recent lpc-815 I just hooked it straight up to an 18650, direct driven.

It probably has 3 minutes on it total being direct driven, in 10-15 second spurts.
With a regular aixiz glass lens, I got 410mW. And it hasn't degraded 1mW yet. So I'm saving it, if it dies it dies. It cost me $5.

What is it that is encased in heatshrink the diode or the driver, it's hard to tell from your description.

the diode is.. to isolate it

So I am totally confused on this front then... hmm. May open it up to see if there is a short somewhere.

Hmm. So opening up must have loosed a short or something, because when I opened it up, I tested the amperage going from the battery to the driver, that was roughly 400mA, and the amperage to the battery was around 405mA. So... yeah. Problem solved.

Your lucky you noticed it running at that current...

My LOC hotlight died in one month running at 429ma

It may have been a measuring error, to be honest. Because it didn't *look* significantly brighter than my 380mA LOC.

It could have been a legit measurement

You wont notice much brightness difference even when you compare a 100mw greenie to a 200mw, thats if they are not side by side

Wouldn't 600+mA kill an LOC nearly instantly?

I had a LOC get accidently set to around 500mA on assembly once, It was running around 330mW with an aixiz glass. It had degraded to around 280mW by the time that host bit it, (was cheap crap. the threads stripped) and I transplanted it into something else with a different driver and promptly LED'd it. The diode had about a dozen hours on it before that happened though.

Fascinating :O

Get the heatshrink off the diode !
find a better way, heatshrinking
a laser diode is not the answer !

Well, it's not on the diode per se... it's on the Aixiz module....

Heatshrink doesn't conduct heat worth a crap heh. I've actually been on the lookout for a thermally conductive but electrically insulating plastic for about a month now. I know people make it, but so far I can't find anyone who actually /sells/ it. At least in a machinable form like rod or tube. I can get it as extrusion pellets but that doesn't do me much good lol.

Yeah, I know. But it was the only way I could think of to isolate it. So I will just use short duty cycles.