Last winter I constructed a couple quad emitter C cell maglites using XM-L2 and XP-L emitters and convoy S series reflectors. Those builds went to plan with some minor setbacks so I though I’d give a quad XHP-50 a try. The basic construction is very similar to last year’s models but I wanted to try a different method of fitting the reflectors into the head. In previous quads I had reamed out the head to fit the reflectors, this time I calculated that I could reduce the diameter of the reflectors to fit the head. This turns out to be a much better look.
There are only three contact points on the reflector when fitting four into a round head. We begin by arranging the reflectors in the crook of a right angle to mimic the desired alignment and marking the reduction points with a sharpie.
The stock diameter is 0.801 inches, the radius of which must be reduced by at least 0.012inches.
Some quick work with a flat file on the vertical axis of the reflector reduces the radius below our minimum value.
Twelve points (3 per reflector) must be filed to allow the reflectors to fit in the head.
They fit!
Four XHP-50 J2 2B emitters were bought from Mountain Electronics and tested with various reflectors and focuses in an attempt to eliminate beam artifacts. The only way I was able to obtain an artifact-free beam was to lightly sand the domes of the emitters and run them under OP reflectors.
The emitters were thermally adhered to a heatsink made of a 1/4 inch thick chunk of 1.875” diameter aluminum rod. Copper strips are used as a low resistance, low profile electrical conductor. The emitters are wired in parallel. I was able to use 18 gauge wire throughout the light.
The driver is a BLF17DD-Z flashed with four mode STAR firmware with off time memory. I’ve been adhering drivers to mag switches with hot glue and I really like how it works without being permanent.
All ready to hook up and test!
Unfortunately, I forgot to thoroughly clean the emitter domes after sanding and three of the four domes showed catastrophic damage shortly after the first in-host test. On the worst of them it appeared that sandpaper particles on the domes had become so hot that they melted the dome almost to the phosphor layer. I was able to save all the emitters by shaving the domes off flat and polishing the remaining material to a smooth finish. The tint warmed some and the beam remained artifact free. Here they are all undressed.
All put back together again under a borofloat lens, I think they look better dedomed.
A couple host mods I did was bypassing the tail spring with 18 gauge wire and boring the battery tube to accept 26650 cells.
And here is the finished light.
This light draws over 15 amps from a pair of powerizer cells and a bit over 12 amps with king kong 26700s. Member 18sixfifty built a quad XHP-50 light with a FET driver and measured around 8000 lumens off 26650 cells. Calculations based on the work of Djozz support that as a reasonable measurement.
Special thanks to member WillyD for gifting me a reflector needed to complete this build.
Happy modding,
Brain
ETA beamshots, I believe the request was for comparative beamshots so here we go.
Fenix TK41 with XM-L2 U2 1D
2C mag, quad XP-L HI V2 3D at 5 amps per emitter, S2 reflectors.
Supfire M6 with stock XM-L2s on noctigon, FET driven, 4x Samsung 26F cells.
Quad XHP-50 2B shaved domes, 12.5 amps, S2 OP reflectors
