Maxwell Render | Official public forum

Hey all!

I dont know why, but I wanted to see if I could achieve a realistic Fibre Optics render?

I've made the fibre optics like this:



The coating has a diameter of 1mm. The cladding is about 0.2mm thick

"A fibre optic cable consists of a glass silica core through which light is guided. This is covered with a material with a refractive index of slightly less than the core. This is called the cladding. The refractive index of the cladding need only be around 1% less than the core to achieve the total internal reflection necessary to confine the light to the core"

I've created the core material and the cladding material.
I've simply started with a Dielectric material with nd 1.5 for the core and 1.3 for the cladding for testing... So far it looks pretty bad

http://itk.ramboll.se/visualisering/render/fibre.jpg
(this is the path im rendering too, so the result will develop during the night..)

This is a printscreen from studio:

http://itk.ramboll.se/visualisering/render/fibre2.jpg

For the lights im using 3 circular emitters, one red, one blue and one green.



And this is a top view of the scene:



Any ideas about how to succeed with this?
Especially I need some help with the materials I think...

The cladding should not have a bore in it when you model it.
It should be solid and the core should float coaxially in it (so that you do not have any coincident faces along the length of the tube)

I understand that's not a top priority, but all the same I wish this coincident polygon issue will be dealt with some day.
HD.

well, you've definately found a sucessful solution to rendering the milky way!!! dont give up on it now.

Thomas An. wrote:The cladding should not have a bore in it when you model it.
It should be solid and the core should float coaxially in it (so that you do not have any coincident faces along the length of the tube)

aaah, ill give that a try!

New try!
New model!





same materials as before...

Milkyway is fun but..

maybe try having a bit of ambient lighting, low so we can still see the effect of the fiber optics--wasn't there a fiber optic render in the gallery of the main site back in the alpha days? I seem to remember that image being a major reason why I bought MW in the first place

There is some problems with the MR gallery... and there is not any more the image you are referring.

You may have a look on this anyway... and there is the same explanation but done with air.
http://www.winosi.onlinehome.de/Gallery_t13.htm

I tryed with MR and it was ok.

the radius of the pipe determines the amount/number of internal reflections, maybe just start with a much simpler version of the test...

this is nearly impossible with maxwell with this geometry... sadly.

* * *

cool test though !

Well I think this looks pretty good.

As above (pretty much) just with a matte light tube to get close to parallel light (well close, also some leakage because my core extends beyond the cladding) using a red bright lamp (non-real data)

60W 5cmx5cm plane above.

20cm of fibre optic.
1.5mm diameter cladding with just under 0.95mm bore, core 1mm... so 0.05mm overlap so no co-incidents but still a change in mediums which pure overlapping doesn't allow (Nd's as above)


Main thing is co-incident won't work, neither will overlapping through core/cladding, BUT having an overlap AND with a cavity appears to work well



Dave

Thats interesting!

Thanks for your imput!

Wow, looks very nice, Mr. Whippy.