Busted Dorcy 41-4297 -- I need your criticisms/suggestions!

I have a Dorcy 41-4297 flashlight (body). It used to have a K2 Luxeon led sitting in it, but thanks to my butter fingers it is no more, so I popped out the pcb that contained the LED.

The above link shows the light fully assembled and its tech specs, plus you can mouse over some of the pics to see the emitter and other features.

I have since ordered a Cree XR-E Q5 star to replace the Luxeon, but I wonder about the flashlight's capability to handle such an LED. You see, the weird thing I discovered is that this flashlight appeared to run without a constant current driver. The pcb which contained the LED had nothing on it except for positive and negative contact points for the 3AAA battery holder and flashlight body, respectively.

Below are images of the PCB so you can see what I am talking about.

I have seen very few discussions about this particular flashlight, even less about its internals. The tailcap appears strange, and I never was a fan of the color-change battery indicator boot. I wonder if circuitry intended for the LED is housed there, but to me it seems unlikely.

Tailcap:


So right now my plan is to get the star when it eventually arrives from DX and thermal epoxy it to a brass heatsink (basically a washer). I will then solder wires from the star to some contact point of my own creation (on the bottom of the star) for + (the center hole in the brass washer will allow for the battery cartridge to make contact with such a point), and send another wire to contact the body of the light for -, along this little ledge here (which seems how neg contact used to be achieved)).

aforementioned "Ledge" upon which pcb once sat:

makeshift brass heatsink sitting in ledge, note center hole for + contact:

My main concern is the circuitry (if any) in that tailcap, which appears difficult to take apart without damaging it. I wonder if it will impede or assist the proper functioning of the Cree led.

I'd say that the chances of the tailcap containing a driver are slim, so the led will most likely be direct-driven. If this is so, I wonder what my output and runtime characteristics will be on 3AAA batteries.

Anyway, I thought I'd submit my plans for hacking on this light to the hive, so please offer your criticisms and/or suggestions.

So it is probably direct drive one mode it should work fine. I personally would have put an XPG or XML in there but an XRE will be a decent upgrade. It should work really well with decent throw. I have made some direct drive XRE lights they pull about 1.5 amps. Here is some good info.

http://flashlight-wiki.com/Brightness_Bins

you don't think an xpg or xml would shorten the runtime drastically?

i figured the q5 xre was a good balance between power and runtime on 3aaa's (yeah i know 3aaa's aren't going to provide much runtime anyway)

Yeah they will kill your run time XPG is going to pull about 1.7 or 1.8 on direct drive tops if that little battery holder can even push that much and the XML would be limited by the battery holder.

I just modded a C-88 and the bottle neck was the 3AA battery holder I couldn't get more than 2.5 amps through it. I think the battery holder will be your bottle neck. If you want a light with throw the Q5 is going to give you the best throw anyways. Compared to what was in there the Q5 is going to look awesome and usually the tint on the Q5s is pretty decent so it should be fine. I just like the 3D tint XPG I have been putting those in any light I can get my hands on the vanilla ice cream tint is so awesome.

Have you taken apart the tailcap? I believe that's where the driver will likely be hiding.

You think the driver would be way aft of the emitter like that? I've only ever seen drivers wired/soldered adjacent to the emitter.

i believe my energizer ultimate precision has the driver in the tail cap. when i did a tail cap reading it started out at 1.some and went up over 4 amps in a few seconds, then poof. but this is also a ramping light though. id check, its a possibility