? for the Driver Gurus(how to build 1A driver)

Hello

*EDIT: The driver described here seems to be useless to me LOL.. I need help to build a suitable driver..

I have the pictured driver from a Coleman 2xAA LED flashlight. The flashlight originally had 5 LED's, each with a tiny resistor marked 1R0...(now it has the module from my KD150 greenie in it [smiley=thumbsup.gif] ).

I also have a CREE 3 Watt LED from the Coleman MAX flashlight(which is now an 803T bluray@150mA).. I have read that the CREE 3W's need ~3.8v@1000mA to run at full power, not to mention a helluva heatsink. The MAX flashlight's driver consisted of one large resistor lol(3AAA power source).

What I'd like to know is what type of driver I have here, and if there is a chance of using it to drive the CREE. The component on the left is a small coil such as I have seen in boost drivers, which is what got my hopes up LOL. I'd like to use two NIMH AA's to power the LED.

EDIT: There are no components on the backside of the PCB. The only pass-thru connections are on the input and output solder points, otherwise the backside is blank..

2nd EDIT: I just did a test using a CR123 to power the driver and the original LED set.. I measured across the + and - ouputs of the driver, LED's lit up of course..I got a reading of 3.78V@2050mA (of course when I tested current, the LED's went out, so I imagine this isn't the way to do it). Is this what I can expect it to feed the CREE?.. if so I think I will have instant smoke... :-/ Measured In-Line with the supply power, I get ~535mA..   Again, not sure what to expect when I hook it up the the CREE..
Thanks in advance

Re: Question for the Driver Gurus

Anyone?

By my results it is obviously a boost driver that seems to feed ~530mA@3.7v from 2.9v source power.. BUT my question is if it can be modified to output 1000mA..

Re: Question for the Driver Gurus

Looks like a "joule thief" style circuit...

Re: Question for the Driver Gurus

It is indeed..similar to many of the boost circuits used to drive our LD's. I just looked up the "joule thief" curcuit on Google. Thanks

Hopefully it can pass 1A of current though.. I am just going to hook it up and see what happens. Hopefully it doesn't overfeed my CREE. Only way I will know is to try I guess..so I will put my DMM in series with the battery to read the current as soon as it lights up, and if it will, hopefully before it blasts the LED..

Re: Question for the Driver Gurus

the driver chip on there is probably a fixed current chip, allowing only 530mA (since I see no other resistors). So yes you could add more current but you'd basically be re-doing the whole driver. OR it could be using the excess current to convert into voltage. Try using a bit of a higher input voltage and see what you get.

Re: Question for the Driver Gurus

Well I looked up a test on this LED.. I don't have to worry about frying it anytime soon..7.84W@1.93A with a Vf of 4.06v .. Too bad we can't get a LD (besides an IR one-although they are still much less efficient) that'll do that!!

I do have the LED arctic silver'd to a chipset heatsink from a PC motherboard though

I jumped the input voltage to 6v.. which kindof kills the need for a boost driver of course. The current going IN to the driver(only way I can measure without making a test load to simulate this LED) was ~1.4 A and the Vf was ~3.45V.

Now according to the chart that I saw from the test results, the LED draws a Vf of 3.45v@~600mA. Are the Vf and mA directly proportional, or is it possible that I was feeding the LED way more current at that voltage? Here is the chart:

CREE XR-E White LED:

Current (mA) 130 310 670 980 1260 1570 1930
------------------------------------------------------------------------------------
Vf    3.15 3.37 3.62 3.75 3.86  3.96 4.06
Watts 0.41 1.04 2.43 3.68 4.86 6.22 7.84
Lumens 42 94 178 234 278 312 341
Lumens/W  103 90 73 64 57 50 44

EDIT: the graph didnt copy right grr.. the values stack one above the other in the order they appear on each line though

Either way I think this driver is useless for this application, at least if I want to run this LED at a decent power..especially since the driver started to smoke LOL. Either way, I would be in need of a boost driver that supplies an ungodly amount of current.. which would kill batteries superfast.

This leaves me with one question..I need ~3.8v@1A to do what I'd like to with this LED. The original supply was 3AAA's with only what appears to be a 1.2ohm huge resistor in between the batts and the LED. The resistor seems to do nothing to knock down the voltage or current. Any suggestions on the proper way to achieve this? I am kindof flexible with the power supply voltage, though I would like to stick with primaries...

The stock setup is only feeding the LED ~3.1V@308mA... They should change the packaging.. at that current it only produces ~90 Lumens, not the advertised 115.. My goal is to push somewhere in the area of 240 Lumens ;D

I am currently feeding this LED with an AC charger from an old cellphone, rated at 5.1V-800mA.. I have the original resistor in series.. The diode seems to be getting 3.26V@810mA..which does NOT match up to the graph above. Only thing I know for sure is that it is blindingly bright :. I am going to have to match it up to a CPU heatsink also.. this puppy gets quiite warm, not hot enough to burn me though.

What I would like to know is what components(hopefully available at radioshack lol) would I need to build a driver(probably DDL-style) capable of supplying 3.5-4.1V@950-1300mA(adjustable) or the 3.8V@1000mA(fixed)? And, what would I have to use as a power supply?

Again, thanks in advance

Anyone? I have done some searching here in the forum since I remember reading about building higher current drivers for c-mount infrared LD's.. but I can't find anything specific on what parts to use

Any help is greatly appreciated since I have some nice ideas for this bright little LED ;D