When looking at it for a glimpse, it reminded me the pattern of a fire salamander 
Great small light indeed
Sweet. Looks like the colour copper goes when its annealed and let cool.
.
Sweet 
Do you have a pic of how you wired of those tiny leds ?
i want that rabit!
could that rabbitbe done to a fw3a?
I’ve become a bit of a tint snob so I changed the emitter on my EDC thrunite T10. The dedomed XPG2 S4 2B was just too green so I looked for other suitable emitters. I wanted to stay with a 2mm^2 emitter to keep the beam size and ability to throw a bit without too much current. Based on my experience and some research I decided to try a dedomed XPG2 1D which I ordered from fasttech. I also switched back to the stock reflector from the TIR I had been using.
I’m very pleased with the tint. It is 5000ish K and very neutral. In the picture below the left beam is a S2 triple dedomed XPG2 S4 2B, the center is the dedomed XPG2 1D, and the right beam is a S2+ with XPL HI 3D. The XPL HI 3D is noticably more rosy. Not sure if the dedomed 1D is above the BBL but it’s very close to neutral.
Olight S1R NW (left):
XM-L2 at about 5000K. Really nice color, previously installed one in the V11R, too 
It’s a FW3A with Lexel’s rainbow aux board.
Comparision with turbo glows:
The biggest gripe was that the aux board does not fit MCPCB installed in my sample.
The hole underneath noctigon triple had to be drilled.
When I was at it, SST-20 FD2 were installed.
Original (not carclo) optic stays for now, since it’s a bit more floody.
Beautiful colours!
I had a bit of a scary experience after modding a S2 with a 2mm WF. I soldered the LED to an XP board and it was fine. I put the light together with one of convoys “ramping” linear drivers. The light was working great and I had been playing around with it for a bit.
I then was adjusting the brightness of one of the modes which with this driver involves ramping the brightness slowly up through the max current which is 7-8A, so the light naturally gets a bit hot during this process. At the end of the ramp it starts again at the low modes. It was here that I realized the light was still getting hot, like very burny hot, as if there was a short, but it was still producing light. I quickly turned it off and unscrewed the tail cap.
I immediately did a very short current test with a clamp meter and it was over 20A, and still producing light. I looked for the usual suspects: reflector shorting the pos LED wire, pos wire cut by spinning MCPCB, with this driver the positive solder point is close to the edge of the driver so I thought that was a potential source.
It turns out the short was at the LED footprint between positive and ground. I measured it the next day at room temperature and it was 20 Ohms. Apparently at high temperature the short greatly decreased in resistance enough to pass a lot of current, but not enough to totally short the LED and shut the light off.
Has anyone experienced this before?
Nice investigation!
I have not had that, but I sometimes feel like when I measure a current, that the output should be higher at that current, a mild short under the led could explain that. But more likely my hopes for a led and the optical efficiency of a flashlight are consistently too high 
Thanks djozz.
Here is the aftermath.
There is black soot/carbon between the positive and ground.
After cleaning it out I saw that part of the dielectric layer was burned, leaving a cavity. I think this MCPCB is probably not usable anymore.
This is the flux paste I’m using. It says solder paste but there is no metal in it, as far as I know. I don’t use a solder paste; I just tin the pads and keep them clean with some flux.
I notice it says to clean to avoid shorts. I’ve been using this flux for years and have had a few similar instances of shorts at the LED footprint, none as serious as this one. Is there a different type of flux you guys know of that would be less likely to produce shorts?
Quick goggling shows 2 kinds of Pro’sKit soldering paste (with pix of each)… acid-free and lots of acid.
Guess which one you got?
Jaxman E1. I changed the switch leds for warm white and replaced the resistor (default 100r for 2k2). Now the LEDs are not so bright, but they can still replace the firefly mode. I also replaced the driver for FET+1 with Anduril so the switch leds have 2 levels of brightness. I left the led wires as they were and used the Anduril files for D4-219c, so the flashlight doesn’t get too hot.
Nice mod, CvP!
I could not find a with-acid variety.
My flux paste is old and filthy looking. I wonder if that has something to do with it. In any case I think I will try a new flux.
Wellp, it also says right on the tin that it’s not for ICs and such.
With: https://www.amazon.com/dp/B00CU7MYS2
Without: https://www.amazon.com/dp/B06X6CYVR4
I just bought some of this soldering flux tonight since I was running out: MG Chemicals RA Rosin Flux Paste.
I have developed a renewed appreciation for convoy S2(+)s. They are compact and the reflector is the right size for medium to close range use. I’ve recently modded a few with regulated drivers and different emitters for biking at night. I use 3A or 1.5A mode depending on the length of my ride and use different size emitters for my helmet and handlebars and whether I’m road riding or mountain biking.
Don’t pay too much attention to the tints in the image because I’ve dimmed the lights to be close in beam intensity to one another, and some of the drivers use PWM to dim while others use constant current dimming.
Top left: S2+, smooth reflector, sliced 3V XHP50.2 1D with convoy linear driver.
Tint of the sliced emitter at ~3A current is close to a cree 3D tint. At 2.95A current I measured 5.5kcd and 70lux ceiling bounce. It has a very wide and prominent corona (due to the excited side-phosphor) extending almost to the edge of the spill.
Top right: S2+, smooth reflector, LH351D 5000K 90CRI with 8x7135 driver.
At 3.03A I measured 5.7kcd and 63lux ceiling bounce. Beam hotspot is very close in size to the sliced XHP50.2. The relative output/efficiency seems impressive given that it’s 90cri.
Bottom left: S2+, smooth reflector, XPL HI 3D with 8x7135 driver.
At 3.04A I measured 12.9kcd and 60lux ceiling bounce.
Bottom right: S2, smooth reflector, 2mm^2 WF 5400K from kaidomain and convoy linear driver. I bought this hoping for a warmer tint, but to my eye it does not really look warmer than the other CW 2mm WFs I have, but it does have a noticeable purpleish tint, so I guess that is below the BBL. At first I was going to change it out, but I tried it outside and I thought it worked ok and provided good contrast.
At 2.97A I measured 23.7kcd and 50lux ceiling bounce.
So each time the emitter area halves, going from 8mm^2 to 4mm^2 to 2mm^2, the throw at the same current approximately doubles. For off-road trails I use the sliced XHP50.2 on the handlebars and the XPL HI on the helmet. The off road trails around here are typically very curvy so not as much throw is needed. I took this setup on the road (actually paved trails) and I felt I wanted more throw because of my increased speed and farther line of sight. Which is why I made the S2 with the WF.
I think those are actually the same product, and there is just a typo in the product title on amazon for the first one. The green in the second one is just a wrapper and they both have the warning about ICs. Anyway I will now heed the warning and hope another flux is better.