PDA

View Full Version : Experimenting with Prismatic diffuser


ROTP
09/29/2017, 06:37 AM
Hi, Im Chris from the UK. I spend a lot of time researching on here and thought i'd post my experiment.
Seeing the results from the BRS video on Philips new light is got me thinking. I was wondering how close I could get to T5 lighting from my Hydra 52's. The idea was to take what was essentially multiple small lights sources (smaller light sources create shadow and reduces wrap) and turn them into one really big one.
The principle is based on my photography experience of lighting where softboxes are used to turn the small flashtube into a large flat light source that mimics window light. My problem with the Philips light is to my mind the diffuser is too close to the light source to have a massive effect.
I bought a 1200mm x 600mm 10 sheet of prismatic diffuser panel and have taken photos in manual exposure mode with a set white balance and shutter speed /aperture / ISO. The results shown are:
Off: No diffuser
Low : Diffuser panel sitting on the top of my mesh tank cover
High Diffuser held 2" from the lens of the hydras

Off, obviously lots of ripple and much more pronounced hot zones on the tops of corals.
High, much less ripple and hot zones seem reduced
Low, almost no ripple, some light loss and hot zones seem much less.

I have to test it long term and as I have plenty of power in reserve am going to increase intensity by 10% on all channels. I make no claims to anything other than I'm trying it, hopefully its of interest to someone and wont see me subjected to a barrage of negativity as my experiments often seem to do on other sites.
Picture shows the differences, excuse the algae not cleaned in a while. A small thing I note is that the black band off the top of the tank is from my mesh cover, its cut short to accommodate the slimline ai brackets and with the low diffuser position the shadow is massively reduced!

ROTP
09/29/2017, 08:03 AM
https://www.youtube.com/watch?v=5zAC_lTLDwA

mcgyvr
09/29/2017, 08:26 AM
yep.. lots of work is going on with the "big dogs" of the aquarium LED industry to attempt to "diffuse" the point source LED lighting..
"TIR" reflectors, newer radion diffuser,etc..
video here of that and BRS since you mentioned them.
https://www.bulkreefsupply.com/video/view/brstv-investigates-a-new-diffuser-for-radion-lights-and-a-closer-look-into-the-ecotech-xr15-led-lights/

People are sandblasting the clear plexi splash shields on some LED fixtures to attempt some diffusion..

However.. many like the "shimmer" of LEDs and you will never replicate the spread of a light source like a T5 which is putting out light down the entire bulb vs light only coming from 6" or whatever circles.. You can get close though..

The phillips light uses TONS of LEDs spread out across a much larger surface.. It has a better chance than a hydra ever will of achieving t5 like "spread"..

oreo57
09/29/2017, 11:47 AM
The puck vs whole area arrays is kind of hindering..

As to shimmer.. yea competing properties..

As to uniform coverage matching Flourescents one only needs to look to our monitors and TV's..

https://www.researchgate.net/figure/274573522_fig6_Fig-6-Design-principles-of-the-engineered-diffuser-a-Ray-tracing-and-b-Energy

Also don't forget that tubes still need to be focused i.e any emitted light leaving at more than say 130 degrees either need to be trained back into that cone or lost..

The fact a Kessil can achieve fairly uniform coverage (at depth) based on a single small "puck" tells you its possible.

As to inexpensive at shorter distances is another question ;)

Point is observations show it is possible.. w/ caveats..

Somewhat ironically (and mostly due to output/cost) most "answers" go in the opposite direction.. to entrain the light perpendicular to the water surface.
And for good reasons of course..

http://spie.org/newsroom/3531-solving-optical-problems-for-led-applications?SSO=1

http://spie.org/Images/Graphics/Newsroom/Imported-2011/003531/003531_10_fig2.jpg

so a "mix" (by distance works, w losses of course i.e o/p) then collimate solution (as to the o/p a little of both mix and collimate) is highly advantageous Except it destroys the chaotic nature of tube-like, which (tubes) has also been trained (reflector efficiencies)
http://spie.org/newsroom/3531-solving-optical-problems-for-led-applications?SSO=1

Some solutions solve things people like. ;)

Just for fun..
http://www.lpipo.com/en/Press_40mmColorMixLens30Degree.html
http://www.lpipo.com/en/Press_200mm250mmFresnelLens.html

Zephrant
10/05/2017, 11:39 AM
I pulled the prismatic sheet (fresnel lens) out of a broken LCD screen. They are very directional though, designed to spread light in one axis. A more general lens would work better I think.

I ended up using the diffuser out of the same panel on my BB light, just taped it on the surface of the light. Helped spread the light in my sump, but the UV bulbs have already smoked little spots in it.

Here it is right after I installed it:
http://www.raisedinabarnband.com/rc/IMG_5654.jpg