A Perfect Dedome?

Has anyone sent a dedomed LED to Craig at LEDMuseum?
(I’m not sure if he’s got a photometer right now, but I’d sure be curious to see a spectrum of the output.
I wonder if the dome is doing any significant filtering of light at the UV end of the spectrum, for example)

Haven't heard but I don't think I've heard of LEDMuseum either. I could certainly spare a T6 to de-dome, got quite a few laying around from all the upgrades.

Hm…
I never took a close look at my de-domed XM’L since it JustWorks™, but this thread got me interested, and…
looks like my emitter, that is working completely fine, is supposed to be royally f**ed?! (look at the bond wires)

Here’s a full pic, not like it contains any extra things to see, but I’m just throwing it in since it looks kind of cool: http://shadowww.eu/blf/DSC_0014.jpg

Edit: took a pic of fully lit emitter (taking a pic of emitter in a ~5A direct-drive flashlight wasn’t as easy as I expected!), seems to look somewhat normal. Hm.

How can that work? 1 wire is MIA, and the other is not connected, and discoloration on the phosphor. Wow, amazin.

That was a truly dirty job, the led should not be working at all, it only has one of the bond wires left ?????? I now officially believe in magic (it seems there's always things left to know about leds)

Unless I'm mistaken those wires are only used for positive, all 3 of them and negative comes from below the phosphor, so that should work just fine (but at lower currents as there will be more resistance).

Hm, maybe bose301s can solve this mystery? After all, he’s a Cree employee :stuck_out_tongue:

Hey, it tests 4.5 - 5A DD with a Panasonic NCR18650PD :stuck_out_tongue:

The missing bond wire must be still attached to the die, but laid over and getting power from the one next door. Each bond wire powers a pair of strips, if the lower two weren't getting power you'd only have 4 of the 6 lit up.

Yes, but technically speaking those strips aren't needed. After all the XM-L2 doesn't use them (and that is how they increased efficiency according to one post I read).

I dunked an XRE in Coleman fuel for 24 hours and the dome came off clean. I popped the ring off by using a pair of diagonal cutters at the base as a wedge. The LED worked fine, but the color temp went from 6000K to over 35,000 K! Looks like most of the phosphor came off with the dome.

coleman fuel?

Oh Oh, it was me who suggested Coleman I'm afraid . Can you make a macropicture of the led?

I am beginning to love pictures of damaged leds, here is one of a damaged, but working xpg2 :

The XRE that I de-domed was actually an XLAMP-7090, the first gen XRE. The led die is covered by the dome material and then a hard plastic lens. It looks like the phosphor is mixed in with the dome material and not a separate layer on top of the die. Remove the dome and you remove the phosphor…

I just reflowed a XPG2 die back onto the substrate(!), and the damn thing works. It got overheated during the first reflow and the middle of the die looked burnt out so I tossed it in the gasoline just for entertainment, found the die no longer attached which is why it acted just like one that's running too hot. Put a tiny tiny tiny little dot of solder paste on the substrate, carefully folded the die back down, and hit it with hot air from the butane torch. And the damn thing works now!

You should ask Cree to hire you, you would lower them the rejects rate :D

P.S. I dunked one XP-G R5 into medical gasoline for about 3 hours but I must have damaged the wires when removing the dome. Damn thing is sooo small, need a microscope to see what am I doing...

I use a 10x lighted loop - it's not too bad with one hand holding the loop, and the other doing the tweezers. Got the loop from FT - they call it 30x but it's really about 10x and 2 LED's in it throws a lot of light. You have to have both - magnification and light, at least for me...

Ill take a led that sits on a mcpcb, put it in a little glass jar with petrol, and a lid. Let it stay there for 4+ hours.
Usually, the dome have fallen of, if not I just “shake” the glass a little bit carefully sideways. When the MCPCB hits the glass, the dome will fall off if its not already off.
So far, I never had to even touch the dome. I always make sure the led/dome points upwards…

I thought to get one of those USB "microscope" cameras! it's easier to work when you have 2 hands at disposal :)

For me in the gas, the dome is always off or lifted, so not a problem with the dome. What I do is remove as much silicone material around the emitter as possible with tweezers without touching or getting close to the wires, but the emitter phosphor surface itself always comes out clean. I'll squirt iso. alcohol all over it before the work and after when all done.

Thanks for all the tips guys. I’m now an official member off the dedoming club after successfully using the petrol (gasolene) method on a poor little XP-G2. Theres a big brother XM-L in the tub as we speak.