[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Light levels, glass lids and reflectors

>Date: Tue, 2 Nov 1999 09:23:44 -0500
>From: Peter_Bradley at hc-sc_gc.ca
>Subject: Light Levels and Glass Lids
>Hello all,
>I have recently started a plant tank and was wondering if anybody out
there has
>any comments regarding what the effect of a glass top on my 55 gallon (LxWxH:
>36x18x20) has on light penetration into the water column. I currently have
>W FL tubes for a total of 120 W of light ( ~2.2 W/gal). I have heard both
>extremes: it has a significant effect and it has very little effect. Has
>actually measured the PAR with and without a glass top? Thanks in advance.
>Ottawa, Canada

It mostly depends on how often do you clean the glass.

At normal incidence losses are about 4% per side (8% total), for diffused
light  losses are higher - about 15-20% (depending on angle of incidence).
My measurements gave me from 10 to 15% losses. Mu luxmeter doesn't have
enough precision. If you think that really need precision I can bring a
digital one and measure, but it deosn't make much sense.
If you glass is dirty losses are unlimited. Thus, try to clean it often.
Also, if 10-15% of light make difference iin your aquarium then your
aquarium as system isn't robust enough (you should have larger safety gap
in lighting). Voltage variation, temperature, lamp output variation, etc
gonna make more difference than glass.

ANyway, I always advocate glass because of many reasons (some of my fish
like to to take a air-walk, I don't like moisture in electrical parts,
etc). Of corse, the choice is yours.

Also, I made some computer simulation of different lamp/reflector
I use my tank (36 x 12" surface) and 24" lamps (each was assumed 1000 Lm
I calculate only lifht flux at water surface without any cdonsideration and
angle of incidence, etc (maybe next time, I'll use more sophisticated
model). No temperature dependance. That meanse that lamp produce same amout
of light even if you put them next to each other. In realty this is not
true. reflector efficiency 80%.I assume some scattering on reflector
surface. The lamp emits light uniformely (each point of surface is
Lambertian source). Light recycling efficiency inside lamp is 10%, That
means if light hits the lamp only 10% re-emits back. Real efficiency
probably even less. I didn't care about illumination distribution at water
surface. By changeing raflector profile one  can create different light
pattern on water surface - uniform distribution, more illuminated area in
front, etc. I use T12 and T5 bulbs. You can estimated efficiency for T8.
Lamps are at 80 mm heights from water

The main results are:
1) all "good" reflectors are within 10-15% of difference. For example,
broad flat top reflector is relatively good, as long as top isn' very close
to the lamps. If top is close to the lamps, then every reflector is bad
(that's why AHSupply gets 60% difference between their reflector and
2) the best reflector looks like Mcdonalds M. Actually, if one negletcs
lamp legths then problem is two-dimensional. There two solutions of this
problem and one is "M" style reflector.
3) it doesn't make any sense to use many lamps.
4) I didn't optimize reflectors for every combination (just used "good
enough" profile), so results can be somewhat improved (5% maximum, I'm a
smart guy:). And if you gonna build a chart it wont' be smooth.
5) it's very easy to make good profile by yourself using AutoCAD or piece
of paper.
Lamp         Number of Lamps    No reflector   Flat top(4) Smooth

  T5           1                   340           720       750       800 (3)
               2                   680           1180      1220      1400  
               3                   1050          1650      1620(1)   1860
               4                   1350          1980      1860(1)   2200 

  T12          1                   340           660       690       750
               2                   680           1000      1020      1120
               3                   930           1260      1280      1410
               4                   1200          1520      1480      1600 

(1) - lamps are close to each other
(2) - lamps are at certain distance from each other
(3) - reflector has 100% efficiency (neglection losses)
(4) - flat top was at certain distance from lamps

if the case of no reflector efficiency is basically defined by the aperture
angle (from lamp to water surface):

u=2*atan(8 cm/15")=124 degree, and efficiency is 124/360=34%. WHen you add
many lamps they start to block each other and efficiency drop.

If someone needs light distribution profile at water surface I can send it.
you a sketch with general principle of "good"reflector design let me know.
However, if you want me to tell what is the best optimal reflector profile
suited for your aquarium....well...it's gonna be more difficult.

Mike, who design weird optics around weird lamp