XHP70.2 dedoming pictures and VIDEO added

I recently dedomed a few XHP70.2s and I really like the results so I want to share with all of you.
Here you can see the stages of the method:

Earlier I only shaved the dome off like on the bottom left but I don’t liked the yellow corona it made.
So I decided to slice and dice as Dale said :slight_smile: And I removed the yellow phosphor around the dies like on the bottom right one.
I have a 1mm thick aluminum plate to help the shaving cut.

The led has two bond wires on two opposite sides as you can see on this pic exposed:

They are ESD purpose and if broke one the led still works but I tried to keep them. You can see the two small yellow dots on the sides. Those are the bond wires under a tiny leftover dome material.

Then need to cut from top next to the dies. Not too close because you can cut in dies so blue light will come out on sides. So I cut 0,2-0,5mm away from dies.

Then cut under the phosphor from sides and lift and scrape off the remaining yellowness. I cut around the bond wires to keep them protected.

Here is a video of the edge phosphor removal (dicing) process:


To test the theory and tint shift, I made a light strip. From left to right: Domed stock led, shaved off dome, shaved dome and removed phosphor from edge.

Fired them on:

Then I putted a piece of paper on them and see what I got:

You can see that the center one is the most yellow and the right without edge phosphor barely have tint shift. The dies have 0,2-0,5mm dome material on top and all sides to protect.
I made 6 of these and then I had a sliced one in my fet driven MF02 so I decided to remove phosphor in that too.

That MF02 made 9060 lumens and 226125 cd with stock dome led.
Then made 7590 lumens and 355500 cd with sliced dome.
The surprise come here because with removed phosphor it made 7830 lumens. Maybe it could easy be a measuring error but good to see it not lost any more lumens. I can’t measured cd yet with final version because it is always foggy or snowy here nowadays but I will edit this post when measured it.
Here is the final led in the MF02:

And here is the beam without yellow corona:

Thank you for sharing that info.

It’s very interesting how much you can change the tint with a few careful slices.

Nice job ZozzV6, thanks for sharing. I have done the the slice and dice but I will be trying the removal of the extra phosphor next go around. The beam looks so much better.

What version did you shave? NW or CW?

I think CW. They come from a dead Imalent DX80

Nice work Zozz, very clean… :wink:

Thank you! The idea came from you;)

I think someone else suggested I try it, to remove the aura, someone that is no longer logging in and hasn’t in a while… but it works, all the flip chip technology emitters benefit from the slice and dice method. (I call it that due to the horizontal slice to remove the dome and the vertical dicing action to remove the excess phosphor)

You do it well though. :slight_smile: Meticulous in the detail department. :smiley:

Thank you for sharing, Amazing skills! :+1: What happens if you break the bond wires on the edge? Does it affect the led in any way? Will the led still work? less output? I don’t think I have brain surgeon hands that some of you were gifted with :frowning:

Did you clean the silicone afterwards? If so what did you use?

Awesome… thanks for sharing the process! Now to work up the courage to try this on a ~$20 LED!

Here’s a thought… what if you could make a cement mixture to surround the emitting area similar to a White Flat?

Scalping XHP70.2 version of NW is an average idea. I convinced myself. This is the version from Thrunite TC20. It’s getting yellow. I’ll take measurements later.

I don’t know the long term affect of the broken bond wire but the 2 I broke work fine at 100mA. They aren’t putted in a light yet so I didn’t tested at higher currents. I just blowed the led with compressed air. didn’t used any cleaning material.

Did you removed all the silicone from the dies top? It seems like that.
I didn’t removed all. I want to see how your turns out ! :wink:

Hmm …. I played with a knife, long and patiently :slight_smile:
Now I want to scrape all the yellow, but it will probably be the diode’s death.

If you remove the yellow phosphor from diodes top they will turn blue.

It can only gather some of that yellow layer?

Nice clean slices. :+1:

The way I do mine is I just take small slices with an exacto knife until the silicone is very thin. As long as the remaining silicone is as thin as possible I don’t think the surface smoothness really matters. With no bond wires to cut it is pretty hard to mess it up. So IMO there is no reason not to slice it. Then again all my lights’ LEDs are domeless (except my headlamp).

This LED sliced is definitely a potent combination of output and throw!

Thanks for the lesson Zozz. :beer:

Hey ZozzV6 did you get a chance to measure CD. of the slice and “dice”?

With dome 226125 cd
Sliced only 355500 cd
Sliced and diced ………?

My MF02 w/xhp70.2 (sliced and diced) lexel’s driver @ 8amps only did 298,XXX CD measured at 9 meters. But my variables of measuring might differ a lot from yours.

Very clean & precisely done ZozzV6!! Thank you for sharing the process & supporting pictures!!! . :+1: