In the general way of things I find myself being a big fan of stuffing large things into small places. When applied to flashlight modding and life in general,this can become an exercise in frustration resulting in angst, destruction, and unanticipated cash expenditure. When I got a 2C incan mag for Christmas and began measuring the space I had to deal with it became obvious that my plan could work and rather easily at that. I should have known better. The plan was to stuff these:
into this:
and power these victims:
With this:
Pretty simple conceptually, more difficult in execution. We begin with a 3/8" chunk of 1.875" aluminum rod shortened to 0.25" thick and narrowed to form a tight press fit when wrapped with 0.012" copper sheet. Drill a center hole and we have a functional heatsink.
We need to increase the ID of the mag head by around 1.5mm to accommodate the overly ambitious reflector selection. Mark with the heatsink installed and gut the head with a dremel tool.
Test fit for width and height. Perfect. Yup,that's two each smooth and OP reflectors for convoy S2. The lens is the 50mm AR coated from KD.
Affix victims with thermal epoxy and wire in 2S2P configuration. The low profile copper strips are needed to clear the bottom of the reflectors.
I was hoping for about 8 amps at the tail with this config so the switch need modding to keep all that current away from the mechanical bits. A wire run from positive battery contact,through a drilled hole and direct to led positive does the trick. Driver negative runs from the negative switch contact to the ground leg of the driver FET. A wire from switch positive to driver positive provides power to the MCU for switching and regulation.
A quick shot of a modified torx tool needed to play with modern mag switches. Looks like the light engine is ready for testing.
Testing proceeded as hoped outside of the mag host. Four modes with a nice moonish mode. The driver kit from Mountain Electronics worked great. I put the light engine and switch into the host, installed the reflectors with what seemed to be appropriate insulation and began by testing tail amps. 10.5 amps of crappy laptop pulls! Then a big spark. Then nothing. A tear down revealed 3 dead emitters and a very pissed off modder. OK my fault. An obvious short under the reflectors. Install three new emitters, alter the reflectors, check for continuity errors and test. Only killed one emitter this time! Happened to be the one that didn't die in the last massacre. OK probably weak from abuse. Excited by such progress, I promptly exchanged the emitter, added some more kapton tape and tested. Killed another one, on the other series chain from the last one.
A consultation thread generated some discussion and inspired by some of Wight's excellent art I altered my wiring by bridging the two copper straps on the far sides of the led configuration. The theory was that a vF imbalance was causing one emitter in the series chain to draw more amps than it's neighbor. Perhaps lethal amps given the recent testing suggesting newer Xm-L2s can't be pushed as hard as the old. The wire bridge is supposed to minimize the negative effects of vF imbalance by...balancing stuff. On the off chance the wiring was still shorting on the reflectors I filed the lip off the underside outer edge of the reflectors and filed down any and all tall spots in the wiring. Then I rather thoroughly insulated everything in sight.
Testing while fully installed revealed full function. It is worth mentioning that while the 2C mag was never meant to accommodate two 18650s in series there is a long history of modders making it happen, particularly for hotwire mods. Simply moving the switch forward by a few millimeters, de-anodizing the tail cap,and installing a low profile spring gives all the space needed for 2S unprotected 18650s to fit. Button top cells probably require a thin conductive spacer at the tailcap/body joint. The solder joint at battery positive needs to be low profile. A small spring can be held in place with a 20mm XM-L insulation gasket.
Finished at last.
Moonlight mode
High mode drawing 8 amps at the tail off Sanyo pink top tool pack pulls. It's 60 yards to the sugar camp. Up close the beam is butter smooth with a very intense hot spot. I'm guessing it puts out 4000-4500 lumens of very nice 3C tint.
That's all until next time.
Brian
