"partial" dedome...

What if i just flatten the silicone dome without removing it completely. Won’t that give me some of the desired effects of a complete dedome with less risk of damage?

The dome is a little too resilient to hold a flat form if you flatten it, or it would probably split if squashed too much, but the main issue would be breaking the bond wires, depending on which LED you are referring to.

What are you de-doming, and how are you proposing to “flatten” it? Some have used a razor blade to slice a dome close to the die…

I’ve used the gasoline method myself exclusively, and have gotten good results from the XM-L/L2 emitters I’ve tried with. I had a bit of trouble with an XP-E2 though, due to the smaller size and spacing of the bond wires.

I inadvertently dedomed a XP-L using the heat method (didn’t realise board had separated from the pill) and that worked perfectly, at least in this case…

XML for now, but wondering in general. Just slicing the dome with a razor close to the surface but leaving the silicone below intact. No heat or solvents. Any thoughts?


Yep, it works fairly well. It’s more popular with emitters that are hard to dedome, like XHP and MTG

Not sure if you’ve already read these threads…

I stand by the gasoline method… I’m worried that using a razor would be more likely to cause damage to the bond wires, or leave a rough surface on the remainder of the dome material, interfering with beam cleanness.

Have done this today (razor) as my first dedoming. Smaller spot, but artefacts and only small improvement in throw. Dedomed more then and also de-phosphored and de-bondet the LED.

I think i’ll try it as my maiden dedoming…

Thanks for the links.


I cant load a pic for the life o’ me!

The dome has functions, some of which are good (protecting the phosphor) and some which we may not want (altering the dispersion of the light). It would be interesting to see the results from slowly slicing the dome away in steps and/or altering it’s shape with testing at each point. Perhaps there is something we’ve missed by working only with the end extremes?