Experiments in LED encapsulation

[Th232]

*BUMP!*

Hi Saab, you're probably busy with making Deft FTPs, but have you had the time to experiment further with these?

 

[saabluster]

*BUMP!*

Hi Saab, you're probably busy with making Deft FTPs, but have you had the time to experiment further with these?

No I have been pretty focused on the DEFTs of late. I have some ideas I want to try so I am not done with this thread. Will update when I can get to them.

 

[BrianMc]

Even if you remove the silicone gel from off the top of the phosphor the phosphor itself is not coming in contact with air. The phosphor is suspended in a silicone elastomer that acts as a binder.

1. Am I right in thinking from the above, that the LED can exist happily (or happily enough) without a dome and the silicone gel if hidden behind a lens and mount?

2. Do you or anyone have a domeless & gelless emitter with lots of hours on it, if so, OK?

3. I have a damaged dome on an XP-G where an XR-E would have been a better choice. Do you or anyone know of anything about an XP-G (other than wasn't the best option to start with) that might be an issue, if I remove the dome and gel rather than replace it?

You may notice that relatively few people come down here to the LED section. It is a place for LED geeks of which group I claim to be a part. Few people here truly understand the magnitude of what's been accomplished here but I'm fine with that. Thanks for noticing.

Awesome. :bow:

Started with LED MR16 plug ins...and in a few months, I'm thinking of modding an LED to get the beam I want.

lovecpf

 

[saabluster]

1. Am I right in thinking from the above, that the LED can exist happily (or happily enough) without a dome and the silicone gel if hidden behind a lens and mount?

2. Do you or anyone have a domeless & gelless emitter with lots of hours on it, if so, OK?

3. I have a damaged dome on an XP-G where an XR-E would have been a better choice. Do you or anyone know of anything about an XP-G (other than wasn't the best option to start with) that might be an issue, if I remove the dome and gel rather than replace it?



Awesome. :bow:

Started with LED MR16 plug ins...and in a few months, I'm thinking of modding an LED to get the beam I want.

lovecpf

Although I haven't done the thousands of hours testing that would be needed to qualify parts to be sold without the outer layer of encapsulation on XR-Es I have not seen any issue in the tens of hours. I have not spent any time tearing down the XP-G however and I am not sure how the phosphor is applied so really have no basis for whether or not it will work without the dome.

 

[panicmechanic]

Started with LED MR16 plug ins...and in a few months, I'm thinking of modding an LED to get the beam I want.

...and you'll be reading stuff like this :naughty:

Pity they haven't analysed the XP-G yet.

 

[BrianMc]

Although I haven't done the thousands of hours testing that would be needed to qualify parts to be sold without the outer layer of encapsulation on XR-Es I have not seen any issue in the tens of hours. I have not spent any time tearing down the XP-G however and I am not sure how the phosphor is applied so really have no basis for whether or not it will work without the dome.

Thanks. Had to ask in the off chance you had.

No point in going where someone has found it to be unwise.

If the domeless XR-E's don't self destruct for at least 10's of hours I might get couple of months on flash in the day, a couple of weeks of night rides. Sounds worth a shot but AFTER I have an XR-E or three on hand, just in case.
:sigh:

Great stuff. I'll be watching.


Anyone else reading this, had any experience running domeless XP-G?
:thinking:

 

[Neondiod]

I think this can be called a "throw job" :naughty:

Do you think there is a chance that you could get a usefull beam from just an LED with the modded encapsulant? That would be awesome to be able to just stick a new dome on an LED and put it behind a flat window for a medium beam and then have the option to throw an aspheric in front of it for a laser.

That's means back to basic 5mm led design.

 

[BrianMc]

A point of clarification:

Although I haven't done the thousands of hours testing ...without the outer layer of encapsulation on XR-Es, I have not seen any issue in the tens of hours. QUOTE]

This infers that you have some XR-Es reencapsulated as per your original post, and some "runnnin' 'roun' nekid" (domeless) with a some tens of hours of run time? No issues means no tint shift either?

Is so, VERY interesting....:thinking:

 

[Fulgeo]

Looking at these experiments a few questions comes to mind. There has to be a difference in the index of refraction between the three materials present here. I would like to think that the acrylic optic lens has a lower number than the dome or the silicon encapsulant. I also notice that the over all height profile of the emitter is slightly shorter with the acrylic optic lens in place. The photons have less material of possibly a lower index of refraction than original. I think you have changed the degree of dispersion is a desirable way. The real scary thing is you got it in your first try. :thumbsup: Have you any information on the index of refraction of the three materials? Is the silicon encapsulant you used as good as the original dome material? Looking at the last two photos in post #38, the projected image does look brighter but also a bit smaller. Just wanted to confirm that the distance was the same between light and target. Thanks for sharing your results, very interesting stuff.

 

[saabluster]

A point of clarification:

This infers that you have some XR-Es reencapsulated as per your original post, and some "runnnin' 'roun' nekid" (domeless) with a some tens of hours of run time? No issues means no tint shift either?

Is so, VERY interesting....:thinking:

The only tint shift I have seen occurs immediately after de-doming the LED. I have not seen any addition shift in my naked XR-Es with time on them. One thing that can be a problem though is that with next to nothing covering the die some pieces of dust that may land on top will burn due to the intensity of the light so close.

Looking at these experiments a few questions comes to mind. There has to be a difference in the index of refraction between the three materials present here. I would like to think that the acrylic optic lens has a lower number than the dome or the silicon encapsulant. I also notice that the over all height profile of the emitter is slightly shorter with the acrylic optic lens in place. The photons have less material of possibly a lower index of refraction than original. I think you have changed the degree of dispersion is a desirable way. The real scary thing is you got it in your first try. :thumbsup: Have you any information on the index of refraction of the three materials? Is the silicon encapsulant you used as good as the original dome material? Looking at the last two photos in post #38, the projected image does look brighter but also a bit smaller. Just wanted to confirm that the distance was the same between light and target. Thanks for sharing your results, very interesting stuff.

I do not know what product Cree uses so it would be kind of difficult to know what was better or worse. I also will be keeping the specific material I used to myself for now. Yes that second shot is both more intense and smaller which was what the aim of this project was.

 

[BillyNoMates]

Just thought I would chip in here after being pointed this way by BrianMc.

I've been running a couple of (unintentionally) de-domed Rebels in one of my bike lights for more than a year. They still work fine, BUT I haven't done any other testing other than checking they still come on and look bright.

For other (intentionally) de-domed Rebels, it is possible, if one is careful, to shear the dome off just above the LED die and still leave a small covering of silicone over the die, so in theory at least, the die should still have some protection from moisture etc. I'll try to get some pictures to show this.

The die uncovered die is 1mm x 1mm, where as the covered die appears ~1.4mm x 1.4mm. In projected images, the uncovered die provides a proportionally smaller image, with slightly better definition - noticeably brighter.

 

[zzonbi]

How much brighter? It's clear that more light will be TIR trapped.

 

[Neondiod]

Hello!

I want to do the opposite. I want to mount a dome on a PLCC6 package to focus the beam a bit. I think of cut of the top of a generic 5mm led and glue it with epoxi on top of the PLCC6. When I saw of the top of the 5mm it will probably be an uneven surface. Do you think this will be a problem or is the epoxi going to fill out the grains and make the package look clear?

TIA

 

[saabluster]

Hello!

I want to do the opposite. I want to mount a dome on a PLCC6 package to focus the beam a bit. I think of cut of the top of a generic 5mm led and glue it with epoxi on top of the PLCC6. When I saw of the top of the 5mm it will probably be an uneven surface. Do you think this will be a problem or is the epoxi going to fill out the grains and make the package look clear?

TIA

It very well may be a problem if you cannot find an epoxy that has a refractive index that matches the epoxy dome. It will reduce the output of the LED. You could use the lens off the top of an XR-E which is glass and relatively smooth on the back side.:shrug:

 

[gcbryan]

Interesting thread.

I don't think this question has been answered directly but maybe it has...would removing the dome from a XR-E and doing nothing else improve throw noticeably in any light with an aspheric lens?

For instance, with the DX flood-to throw that everyone is familiar with what effect would just removing the emitter dome of that light have?

 

[saabluster]

Interesting thread.

I don't think this question has been answered directly but maybe it has...would removing the dome from a XR-E and doing nothing else improve throw noticeably in any light with an aspheric lens?

For instance, with the DX flood-to throw that everyone is familiar with what effect would just removing the emitter dome of that light have?

Yes it should improve throw. Maybe... The problem with the XR-Es is getting it to pop off cleanly. All too often the gel will tear and then you start losing lumens fast like. I would guess you could have up to a 40% increase in intensity if it was removed perfectly. Just a guess.

The end result of this in a reflector would be a smaller and more intense hotspot with a dimmer spill.

 

[fdrk]

Hi everybody, very interesting.
Can someone tell me a good and not expensive optical grade silicone?
fdrk

 

[CKOD]

Hi everybody, very interesting.
Can someone tell me a good and not expensive optical grade silicone?
fdrk

None, not trying to be a nay sayer, but Ive looked long and hard and couldnt find anything, its pretty much dow corning or bust. and the dow corning silicone is $400 for 1L part A, and $400 for 1L of part B IIRC. Ive toyed with some optical epoxy from expoxies.com, and it was looking like I was going to have CTE problems, or long term bonding problems.

 

[saabluster]

Hi everybody, very interesting.
Can someone tell me a good and not expensive optical grade silicone?
fdrk

You can have good or not expensive but not both. Here is some that is as cheap as you will find anywhere.

 

[CKOD]

You can have good or not expensive but not both. Here is some that is as cheap as you will find anywhere.

Hey, thats pretty new, I was all over smooth-ons website looking at other products for the longest time. Neat find, wonder how it compares in layers that arent just a thin coat.