Easy In-situ Light Pipes

[JNCarr]

I'm making my daughter a Christmas present that requires the light of 110 LEDs to be brought from pockets in the back of the facepiece to the front. The holes for the light are 1/8" diameter.
I looked at acrylic rods inserted into the holes and various materials that could be injected, including silicones, acrylates, pourable epoxy, etc.
After about 60 tests, I settled on hot glue made by Adhesive Technologies. It is colorless and has a refractive index of about 1.48 which closely matches the epoxy used on the LEDs. This reduces internal reflection and offers better light transmission. It's available on amazon. The pourable epoxy would be a good material choice, but its low viscosity makes it a challenge to contain in this project. Also the other materials offered an advantage of over injecting (from the back) by about 1/8" to form a small dome. This gives a wider field of view for the light. The other advantage is it cools in a few minutes and is ready to go.

The LEDs emit at about 160 degrees, so I included a chamfer at the inlet to help capture more light. Shown in the rear picture. The top row is the glue.
Maybe this will come in handy for someone else.

 

[wndopdlr]

Thanks for the info Joe. Any plans for you to show us the entire project when you are done?

 

[Echd]

Interesting, I'm curious what the final result is to be like.

You're smart to be wary of the heat. Even LEDs that are relatively cool can die quickly when cooked... and if they're retained by poured epoxy, they're impossible to service.

 

[bainin]

Modern LEDs do not use any epoxy in their optical path. They are encapsulated with Silicones because epoxy has long term discoloration issues in conjunction with heat and blue light.

I'm not sure I would put epoxy or EVA (hot glue) in that light path unless I felt that the actual light load/on time would be low. I don't know if EVA is also susceptible to browning/discoloration but since I haven't seen it in my career path..I expect it does.

Matching the index of the LED silicone reduces (epoxy/silicone) interface back reflections but leaves you with a strong final exit index contrast (epoxy/Air) which will be a larger loss.
I would instead use a lower index final step instead and curve it like a lense for better optical extraction.

A medium/high shore silicone with a 1.41-1.45 index would be my choice.



b

 

[JNCarr]

Thanks for the info Joe. Any plans for you to show us the entire project when you are done?

Yes - I'll post the project build.

 

[JNCarr]

Interesting, I'm curious what the final result is to be like.

You're smart to be wary of the heat. Even LEDs that are relatively cool can die quickly when cooked... and if they're retained by poured epoxy, they're impossible to service.

As indicators, not high-powered ambient lighting, these run at relatively low power and I've estimated the heat rise to be only about 10 degr C over ambient which is indoors.
Yes - will post the final project

 

[JNCarr]

Modern LEDs do not use any epoxy in their optical path. They are encapsulated with Silicones because epoxy has long term discoloration issues in conjunction with heat and blue light.

I'm not sure I would put epoxy or EVA (hot glue) in that light path unless I felt that the actual light load/on time would be low. I don't know if EVA is also susceptible to browning/discoloration but since I haven't seen it in my career path..I expect it does.

Matching the index of the LED silicone reduces (epoxy/silicone) interface back reflections but leaves you with a strong final exit index contrast (epoxy/Air) which will be a larger loss.
I would instead use a lower index final step instead and curve it like a lense for better optical extraction.

A medium/high shore silicone with a 1.41-1.45 index would be my choice.



b

Thanks for the thoughtful comments. I had looked into the activation energies of the various materials and think the EVA will be fine at the given thermal and flux loads. You would not see EVA in the LED industry because of its very low melt temperature - it wouldnt take soldering operations in board assembly.
As for refractive index match, I agree that some sort of grading would be ideal. But if the LED manufacturer terminates the optical path into air at 1.4-1.45 I would think I am no worse off by picking it up at that value and terminating it into air. Maybe I'm missing something. Yes a hemispherical ("curved like a lens") shape is what I'm referencing by "dome".

Please PM me if you'd like to get into more details - happy to discuss further.