Looks very much like it, but there are also quite some differences (I looked at pictures at manafont.com), the most obvious one is the body of that Small Sun is not one piece. Anyway, it is the reflector that does the job 
.
Successful gasoline de-dome on XP-G2, as well. Working just fine at 2.8 amps. Previously mentioned resurrected XM-L2 U2 working fine at 3.8 amps.
I'm sure there are tools made for this, but I don't have one, so... I made a plastic pointy-thing from the tail end of a zip tie, cut to a very very fine sharp point. Use that to clean away any of the crumbly bits of silicone off/away from the dome. (I used a metal knife on the assumed-dead XM-L2 in the earlier picture, hence the nasty scratches on the substrate - no more of that, thanks.)
Has anyone tried kerosene on these?
If somebody wanted to make a blue one, on purpose, what's the best way to get rid of the phosphor layer?
Finally got around to cutting the grass and since the mower needed gas the XRE started it’s bath late today. It’ll probably soak until some time Thursday morning.
What kind of current are you thinking? Or have you already done it?
The ones I've done in gas only need 3-4 hours, even without giving it a headstart by lifting the corners.
Well, I pulled it out of the dunk this afternoon and there wasn’t any sign of remaining silicone on the ring or the mount. Used bent needle nose pliers to pull the ring off (green stuff adhered on bottom of the ring, spots show in the pic below). Sprayed it off with some electrical contact cleaner to dry it off fairly quickly.
Results
Applied power for a just a brief moment and it still lights up. So I guess I’m now 1 for 1 on a successful de-dome.
Perhaps swapping out the XPE in a single mode SK68 to see what happens is the way to go.
A reflow over to a copper sinkpad and into a P60 drop-in with a new driver would be nicer. Will 2.8A be too much, or should I just limit it to 2.1A? Suggestions on how much amperage to reasonably push into it?
There's no SinkPAD for XRE's?? For XRE's, limit is about 2.2 - 2.4A or so, I've used 2.45A for A60's, maybe not much more than 2.1A. I'm using XP-E2/SinkPAD's now in A60's at 2.8A -- Match's tests still show a rising lumens output at that amperage, but not rising much...
Good question - I think the MTG and XML pads are too narrow, and Sinkpad doesn’t seem to list it.
depends on your XR-E. people here have gone north of 2.5A with white R2s, and that’s without copper PCB.
I can say from experimentation that the green ones are not as robust. Also, the Vf on greens seems to be much higher than white - ie I can direct drive a green xr-e off one CGR18650CH and it won’t eat over 1.4A because of the Vf, but that doesn’t seem to be (as much of) a problem w/ white xr-e
I sent an XR-E R2 to Match and there’s a graph of his test on here somewhere…
edit: also, I flowed an xr-e to a copper sinkpad, and w/o an good tests to back it up, I’d say its a toss up between using that and an aluminum pcb (w/ proper xr-e solder pads), because you lose so much contact area on the center pad.
nope, I asked them directly.
and since it isn’t something they already make, they won’t mix it into an order like they did with out xml/xpg orders.
I just did my first on-purpose xml dedome while the led was in place in flashlight. It worked great. Did it as follows:
- I lifted the led base off the pill so it would have no heat sinking. Turned on the light to high. This is a cheap zoomie with stock driver. High is less than 2 amps, probably around 1 1/2.
- I darted into another room to grab a sharp pick. Took probably less then 20 seconds. Smoke was coming off the led/base. I turned off the flashlight. Took the pick to the dome and the led slid right off the pad. I quickly slide it back on to the pad. I took advantage of the accidental reflow moment to push the led down and squeeze out the excess solder. I then gave it a few seconds for the solder to solidify.
- I then carefully stuck the pick fairly low on the doom on the opposite side of the bonding wires. I put gentle upward pressure. I could see an air gap starting to form in one corner of the led, but I could sense the led was cooling because the doom felt like it was stiffening. I turned on the light to heat it up. When I started seeing a little smoke again, I turned it off. Carefully made sure I didn’t reflow again. Lifted the doom more. I was very cautious and repeated turning on and off a few more times, but only for a few seconds (before it started to smoke).
- The dome came off clean off the top of the led. No residual silicone on the led. The phosphorous looked perfectly intact. Very clean looking. The silicone sheared off above the bonding wires. Perfect, I wanted them to still be protected. There was also residual silicone on the led’s pcb around the led. I left it in place. Pushed the led base back on to the pill and reassembled the zoom and lens to the flashlight.
The light worked great. Reduced the size of the flood, but beam is more intense and warmer as expected. Image of led zoomed out is clearer. Then I got over confident and picked off the residual silicone that was around the led. I did this whiled the led was cold. It peeled off easily. But apparently, it was still bonded to the edge of the phosphorous and pulled a couple small chunks of phosphorous off. You can’t tell with the naked eye, but when the light is zoomed out, you can see the missing chunks in the projected image of the led. Not a biggie because I don’t use full focused zoom in real life. Too distracting. I take it just out of focus so I can see what I am actually trying to light up better.
I have done it with slight cutting then heating and pulling off with tweezer some times and it works every time. The dome is clear and the bond wires still a bit messy…
Is this as easy with an XML2?
There may be an even better way. All you have to do is build a rocket to launch the LED to fly close to the sun to heat up the dome, then a super-intelligent robot can poke it with a pin and peel the dome off. Do you think that might work better than just dunking the damn thing in gasoline and then a few hours later like magic you have a freshly de-domed LED?
Been there, done that. Problem is robots eats phosphors…
I get the impression you may be mocking me. That’s cool. It’s all good with me. I’m telling you though, the led was clean and probably took less than a minute to dedome (not counting other issues). No left over gunk like I see in most of the pictures using gasoline.
Have you tried it? What's left cleans off very very easy. Just crumbles away. A spritz of rubbing alcohol from a spray bottle gets rid of most of it. Anything left after that falls right off with a little brush from a pointy piece of plastic.
I think most of the 'after' pics of the gas method are taken before it's been given a final cleaning, which might be what's throwing you off.
No, I haven’t tried the gas method yet. I wasn’t trying to dis that method. It sounds like a good approach.
I was just being lazy and in a hurry. Turned out it worked quite well. In hindsight, it makes sense because the hottest spot is the led itself. The thing you want to get the silicone off. Was just sharing my experience with a method I haven’t heard about yet. This thread seemed like the best place to share it. If it wasn’t, I apologize.
No no, it's just that I tried (most) all the other methods too and they were miserable failures, not a single one of them worked afterwards for more than 3-4 seconds before going POOF. I've done 5 or so with gas and all have been perfect.
Thanks for sharing ImA4Wheelr. I’m a gasoline dunker myself, but the best method is what works for you.