[ivox3]

Commendable work lebbeus .....and Glebe ---- what can I say ? ....super stuff, but no surprise.

My submission is till forthcoming , ......just been tied up.

__________________________________________________________

Mark, ....good point. I think at the end of this study (and it's nigh), ...we'll have solutions for both. Yeap.

[lebbeus]

thanks guys

One thing that bothers me... is that since LEDs are so tiny and standard applications will be used only as a light source... using IOR and a all that jazz will add a sure unecessary render time.

And since there are usually two kinds, We could have a distinction between opaque (difused lens) led and semi-transparent led.

the material doesn't use complex IORs, it's basically the typical plastic material, just inverted (lambertian layer is less than 100%) and made transparent. It also clears up much faster than trying to "cheat" and use sss…I am open to suggestions for other optimizations--does using a lambertian layer increase rendering times?

A good start for the diffuse lens-type would be to swap the values for the lambertian and shine layer (lambertian 100%/shine 60%?)--this was how I was originally trying to make the material, but was getting frustrated because of the lack of crisp caustics…

I'm working on figuring out colored emitters, I can get a good color if the color of the emitter is the same as the color of the epoxy, but I'm not sure this is correct for modern day LEDs--seems like the epoxy is one standard color and the emitters change, but this might be an illusion…suggestions anybody? Glebe did you use a white emitter or colored emitter?

I had a test going last night for a traffic light, but after 8 hours it still wasn't resolved--it's made up of 118LEDs and a fresnel lens…the file was almost 400MB, don't think it's a viable method for making a traffic light…

this has been fun (now if only I can get that damn retroreflective material to work…)

[glebe digital]

Glebe did you use a white emitter or colored emitter?

Pure white, adjusting colours in the epoxy for this initial 'run of the mill' LED.

And yes, no IORs required.........so far.

[lebbeus]

this was a test to see if the roughness (body glow) layer needs to also be transparent--same as the previous test, lambertain layer set to 60%, however this layer had no transmittance. This image rendered for 24hrs. The preview showed the potential for some glow, however not nearly as much as the other test and it did not resolve as quickly

[glebe digital]

Yes it's not really happening there, but a commendable test all the same lebbeus.
I think the original 'dual-layer' approach will work very well for the basic diode, what other applications should be considered now?

[Thomas An.]

Is real epoxy actually that cloudy ?

(to answer my own question)
I guess it is as cloudy as they want to make it:
http://seniordesign.engr.uidaho.edu/200 ... 0Green.jpg
http://aduro.smugmug.com/photos/61002729-M.jpg

[ivox3]

It's never cloudy Thomas. Unless it's a particular type with additives used for special adhesion ------ plastic and metal epoxies being the ones that come to mind.


EDIT: It maybe have a certain coloration(yellowish usually) that comes from UV exposure, .....but it remains, ....optically clear.

[Thomas An.]

It's never cloudy Thomas. Unless it's a particular type with additives used for special adhesion ------ plastic and metal epoxies being the ones that come to mind.

EDIT: It maybe have a certain coloration(yellowish usually) that comes from UV exposure, .....but it remains, ....optically clear.

Right, I see ... I meant to say "silicon" though (what they use for LED lens).
Sorry for the typo

[markps]

Ok.. here are a few refference photos I've just taken... I'm sorry for the blurry pictures... It is a pain to take pictures of LEDs

Here several different levels of lens transparency.





On this one the light is very difuse... I would imaginge the solution would rely heavily on SSS



This one is not as opaque..






This tiny triangle LED is very interesting... it is opaque on the tip and transparent on the body.





This one is very transparent. You notice that not so much light escapes from the sides.





This one is very interesting... It has a clear lens and the lens it is on a more closed angle so even less light escapes from the sides and you only see the light when directly facing the LED.





Cheers!
-Marcelo

[NicoR44]

COOL thanks!!

[glebe digital]

Great reference shots Mark, thanks.

[markps]

Here you see the light pattern on a blank sheet of paper. The light pattern of the semi-transparent LED is much different than of a Clear LED.. (even though the picture I've taken is not on the dark )





Clear lenses... eidit: the following pictures I've found on the web.

[Thomas An.]

Excellent reference images. Thank You !

[markps]

Even though you can use any kind of LED for anything... There are a few specific epoxy lenses for each application....

Non-Diffused (clear)
Clear LEDs are used when the application requires the strongest light you can get with no need to worry about the direct view of the user. This includes high intensity LEDs and white LEDs. Examples are flashlights, backlight.

Many of the indicator lighs on enclosed equipments come from lightpipes re-directing the light from surface mount LEDs. Usually surface mount LEDs use clear lenses. Indicators and displays can use clear LEDs with the addition of Difused LightPipes or difused lenses.

Non-Diffused LEDs have a much narrower viewing angle and light is focused about +/-12o from the center.

Tinted

Tinted LEDs are used to indicate the color of the LED when the light is off. A good example is the indicator of power... even when the power is off you know the indicator is RED so you expect it to be lit when the power is on.

Diffused

Diffused LEDs are used as indicators or displays. Whenever the user will be looking directly to the LED. Examples are Segment Displays, Matrix displays, indiacator ligths.

The viewing angle of the Difused LED is very wide and in most cases reaching aboove +/-35o from the center. The viewing angle changes quite a bit depending on several factors like the shape of the lenses, Brightness, wavelengh.

Viewing Angle, wavelengh:
http://www.theledlight.com/technical1.html
http://www.theledlight.com/technical3.html

Viewing Angle, Brightness:
http://www.grandwell.com/vw_angle.htm

[lebbeus]

from the tests I've been doing, it seems to me that both the "optically clear" and the "diffuse" can be accomplished through the two-layer method, just change the weight of the "body glow" layer though I don't think going lower than 10 will be helpful…

I had been wondering if some LEDs had multpile surface finishes on the same LED, thanks for the images and links