Eggcrate for focusing light

[Brian Ward]

Did anyone else try putting the eggcrate over their tanks to focus the light the way Anthony Calfo suggested? I did this - tried it both ways - and see no difference. If anything, I think the light is more diffused. Again, tried the eggcrate both ways since 1 way is supposed to diffuse while the other will focus straight down. Maybe the eggcrate I bought isn't designed the same way?

[extreme_tooth_decay]

Hi Brian,

 

I spent some time thinking about this after hearing Anthony talk about it.

 

There are so many variables (height of light, type of light, intensity of light, width of light, width of tank, height of tank, etc) that could effect this that I'm willing to believe that for some setups there may be a net gain, I'm positive that there will not be a net gain for all setups.

 

For my tank, I have a narrow strip of PCs at the top (close to the water), and a fairly shallow tank. The light just doesn't fan out enough to have much loss through the sides in this particular instance. I'm sure I would have net loss.

 

I would like to see some hard PAR evidence with testing on several different setups (light close to surface, far from it, MH, PC, T5, wide tank, narrow tank, tall tank, short tank, etc).

 

just my $.02!

 

tim

[Origami]

Hi Brian. I just received a PAR meter that I ordered as part of a group buy and have been thinking about running this test with white egg crate. This is as good a reason as any to break it out and test it, I guess. I would make my measurements over a grid and through air rather than water in order to standardize measurements and to minimize interactions and confounding. Provided my test set up is constant (and why wouldn't it be, I would be doing the two or three test configurations in sequence without modifying the core setup), the readings should be good enough to answer the question, I'd suppose.

 

Besides white egg crate, I recall having seen metalized egg crate while on a visit to Piedmont Plastics last summer. Given that it's a little more reflective, I wonder how it would fare in this application and if there would be any significant impact to spectrum?

[Coral Hind]

Tagging along as I have thought about this several times since the meeting. I am really interested in seeing some test data too because I would think the grid would block more light rays out instead of focusing it.

[zoozilla]

Tagging along...

[lanman]

His premise was that not as much light would spill out through the sides of the tank. So... you will need to test over a wide area of the tank, including the edges.

 

bob

[Almon]

Yeah, he said less light leaving the tank and a net gain of (more) light in the tank. I'm very hesitant to believe that's true. Let's see the statistical data...

[jason the filter freak]

Even it there is no net gain as long as there isn't net loss it may be a good idea just too keep critters in the tank and undesirables out of the tank.

[extreme_tooth_decay]

Type of reflectors would play a big role here as well.

[scott711]

interested in seeing some data, as I asked him the question about light loss vs. light gain. As mentioned above, the goal was to focus more of the light down into the tank and have less escape out of the sides. I bought some eggcrate to try it out since it would also help keep fish in the tank

[Origami]

His premise was that not as much light would spill out through the sides of the tank. So... you will need to test over a wide area of the tank, including the edges.

 

Hence testing over a grid.

 

Type of reflectors would play a big role here as well.

 

Agreed. I figure that there are at least three categories of light that escape. The first is light that passes through the gap between the lamp and the tank. This lost light would not be affected by egg crate placed on the tank top. Another category of light would be light that leaves the lamp, passes through water and exits through the sides. Some of this light may, in fact, be useful as it may illuminate photosynthetic inverts somewhere in the path of the light on the way out of the tank. Egg crate may either improve this by redirecting this lost light into the water column, or it may reduce this through blockage. The third category of light that exits the tank is reflected and should be minimally affected by the presence of egg crate. The introduction of egg crate creates a 4th category - light that reflects off of the egg crate and is lost before making it to the water.

 

I also figure that the geometry of the bulb, crate, and target will figure into the outcome as well since blockage by the egg crate will be affected by the test setup.

[astroboy]

I checked out the egg crates I use on my two aquariums and all happened to be thick side down. I had a look at thick vs. thin side down. Overall, its hard to see any difference. However, I could see that a few dark spots in the tank, such as the corners and some nooks and crannies between the rocks were significantly better lit with the thin side down.

 

I need to know a little optics for my job, and a few calculations indicated that having the thin edge down would tend to channel the light downward, as Calfo noted, so on paper it seems reasonable. I found a website somewhere that cites him as saying that thin side down increases brightness by 25%, thick side down decreases it perhaps 10%.

 

The eye detects light logarithmically. That is to say, even though the sun is 400,000 times brighter than the moon, the eye perceives it as only about 13 times brighter. An increase in brightness of 100 is seen by the eye as a 5X increase in brightness. This allows you to see pretty well in both very dim and very bright light. The eye is quite a piece of engineering.

 

A 25% increase in brightness would be hard to see, unless you were doing a side by side comparison. On the other hand, channeling light straight downward would likely light dim areas between the rocks quite a bit and you'd probably be able to see that pretty well.

 

Having said all that, it seemed to me my tank was overall dimmer with the eggcrate than without. Obviously, the eggcrate is going to block a certain amount of light since its rather opaque, perhaps 10%, so you've lost some right there. However, without the eggcrate less of the light is directed straight down, consequently, some of the light towards the front of the aquarium will be directed backwards, both from the bulbs and reflection from the front glass. This would tend to illuminate the face of the live rock more than just having the light directed downwards. So, your live rock wall might get more light without the eggcrate, but at the expense of the dark corners.

 

So... Calfo might be right. It would be interesting to see some actual measurements, including the amount of light coming out the front of the tank.

[davelin315]

I flipped some egg crate over at school today as it was "wrong side up". That said, my thoughts are that it will help to focus light if it's placed below the light versus on the tank. For instance, I use Lumenarc 3s and feel that if I put it directly below the UV shield it would help it to focus ligth down, but I also think that I might be compromising the amount of light that leaves the light itself. As it reflects, it degrades bit by bit, so I think overall you wouldn't lose light to the floor, but you would lose light to the egg crate. No statistics to bolster this, just conjecture.

[jason the filter freak]

Wonder if thats what the goofy looking "grill" on the coralife HOB metal halides are for?

[Origami]

OK. I hadn't planned on doing this tonight but I couldn't resist. I performed six single run experiments across two basic configurations (three tests each), taking 9 data points in each experiment. A total of 54 data points were collected and are presented and plotted here. The configurations included both a compact fluorescent and a metal halide configuration. Tests performed were: a) no egg crate; b) egg crate pointing up (thin edge up); and, c) egg crate pointing down. Where egg crate was used, a cut-to-fit piece of white polystyrene egg crate was installed at the top of the aquarium. The lamps were mounted 3-4" above the aquarium top. Measurements were taken using my new Apogee PAR meter. Here are the slides:

 

Apogee PAR meter used to take measurements. (I removed the screw on the bottom of the sensor so that it sat flat at each test point.

 

I elevated a spare 20-High tank that I had (by suspending it over 4 water glasses) and took measurements at 9 points under and outside the tank. Test points were horizontally and vertically symmetric around a central test point as shown here:

 

Here's a picture of the test points, marked by pennies on top of a pool table:

 

Two configurations were screened. The first configuration used a Current USA Orbit, 2x65W CF fixture equipped with dual actinic and dual daylight bulbs. Only the dual daylight bulb (6700K / 10000K) was powered. Since the bulbs, collectively are centered under a single reflector, I expected a slightly unbalanced distribution of light from the fixture with a bias to one side. Since this condition existed in all tests, the control experiment (no egg crate) should establish the baseline from which other readings could be compared. The fixture was turned on and the bulb was allowed to warm up for 15 minutes before data points were taken. Data points were taken over about 5 - 10 minutes.

 

The second configuration used a 175W 14000K Hamilton MH Mogul powered by a blueline ballast and mounted under a polished aluminum parabolic reflector. Josh Yauger should recognize the DIY fixture that I got from him in October. This setup was a little more complicated as the light had to be supported by a makeshift frame taken from a spare Gorilla rack that I had nearby. The rack was set up in such as way as to not interfere with the light distribution to the test points. The light was powered and allowed to warm up for 20 minutes before measurements in order to allow the bulb output to stabilize.

 

Here's the tabulated data. PAR measurements are in units of micromoles photons per square meter per second.

 

And here's the data graphed.

 

Note that in each experiment configuration, the control case where no egg crate was present delivered the most PAR to the test points beneath and outside the aquarium.

 

My conclusions from this rather simple test are as follows:

• 
   
  
• Egg crate blocks more light than it focuses in the direction we are interested in (that is, into the aquarium).
  
• Egg crate does reduce the light that spills out from the aquarium through the glass panels.
  
• Measurement deltas between EC Up and EC Dn measurements are extremely small. Further runs would be required to determine if there is a statistically significant difference between the two.
  

 

I found it additionally interesting to see how the PAR from this little CF fixture compared favorably with that from the 175 MH. Granted, it's a 6700K/10000K lamp vs a 14000K lamp, but it's still pretty good.

[jason the filter freak]

Wow way to take the initiative on this neat little experiment... now what about instead of covering the whole top of the aquarium just have a couple/few inches of egg crate rimming the perimeter?

[Origami]

While it would probably decrease the light thrown outside the tank through the glass, I don't think it would contribute much additional to the light going into the tank. It seems that there's a significant amount of light absorbed or scattered by the egg crate. One thing that you will notice from the graphs above: A good reflector makes the most difference. If you compare the No EC case for both the CF and the MH configurations, you'll see that the CF reflector tightly focuses the light into the aquarium while the MH reflector has more leaking out the sides. Some tweaking of the MH reflector to capture and direct the light could make a real difference, I suspect. But you've got to remember, the CF fixture was made for a 24" aquarium and was, therefore, probably designed to focus light into a 10-13" deep (as in height, width, depth) aquarium, while the parabolic reflectors for the MH fixture were probably bent to focus light down into an 18- to 24-inch deep aquarium.

 

Summarizing: I think a good reflector does much more than the egg crate. If you want to make a real difference, get a good reflector. You can deliver the same PAR at lower wattage or deliver more PAR where it counts for the same wattage.

[Almon]

Excellent results. Eggcrate reduces the light in your tank. This is what I would have expected.

[Coral Hind]

Awesome job! That is exactly what I thought would happen too.

 

I remember Dana Riddle mentioned the benefits of eggcrate to focus light back in a FAMA issue around 2000. Everyone tried tests at home and came up with the same conclusion that you just did.

[rocko918]

funny, i did this 2 years ago. not to the same extent but got similar results. I don't use eggcrate anymore.

 

http://www.wamas.org/forums/index.php?show...amp;hl=eggcrate

Edited December 11, 2008 by rocko918

[Brian Ward]

This is also what I expected logically, but coming from what I would consider a well-respected speaker, I thought maybe it would result in counter-intuitive results. Thanks for doing the work!

Edited December 11, 2008 by Brian Ward

[extreme_tooth_decay]

This is also what I expected logically, but coming from what I would consider a well-respected speaker, I thought maybe it would result in counter-intuitive results. Thanks for doing the work!

 

 

Thanks for doing this.

 

 

I agree these results are not surprising, but I would not be surprised if there are some (possibly unusual) configurations where there may be some (negligible) net gain. There are so many variables, it's hard to test them all. Your results show a very clear difference, though. It's not really even close.

 

 

This just goes to reinforce my underlying fundamental philosophy on this hobby:

 

Don't just believe things people tell you. Always test to see how it works for your setup.

Demand evidence. If it does not exist, be even more suspicious. Try to make it yourself.

 

I can't tell you how many times I found that "conventional wisdom" just didn't work for me or didn't result in noticeable, measurable difference.

(For those hankering for an example, I don't want to argue about it, but recirc skimmers is a good example. I didn't see the money on my setup.)

 

tim

Edited December 11, 2008 by extreme_tooth_decay

[astroboy]

OK. I hadn't planned on doing this tonight but I couldn't resist. I performed six single run experiments across two basic configurations (three tests each), taking 9 data points in each experiment. A total of 54 data points were collected and are presented and plotted here. The configurations included both a compact fluorescent and a metal halide configuration. Tests performed were: a) no egg crate; b) egg crate pointing up (thin edge up); and, c) egg crate pointing down. Where egg crate was used, a cut-to-fit piece of white polystyrene egg crate was installed at the top of the aquarium. The lamps were mounted 3-4" above the aquarium top. Measurements were taken using my new Apogee PAR meter. Here are the slides:

 

 

 

Thanks for doing this. These are very interesting results, however, there's a joker in the deck which we might need to think about.

 

For a ray of light traveling through air and hitting a pane of glass at 90 degrees (that is, the ray of light travels perpendicular to the glass), about 4% of the light is reflected. Since some of the light that goes into the glass is eventually reflected out again, after hitting the inner surfaces of the glass 1-many times, the total amount of light reflected back is about 7.7%.

 

For a ray of light traveling through water, the total reflected back is only .7%. The website http://www.phy.ntnu.edu.tw/ntnujava/index.php?topic=205.0 has a good visualization of this. n2 is the index of refraction of the glass (1.5), n1=1 for air, n1=1.33 for water.

 

 

The math gets complicated for angles other than 90 degrees (google Fresnel equations), but the the 7.7 to .7 ratio is going to stay about the same, regardless of the angle at which the light hits the glass.

 

The point here is that for these measurements, where the aquarium was filled with air, a fair percentage of the light hitting the glass was reflected back into the aquarium. The effect of the eggcrate to focus the light downward (and off the glass sides) is sort of negated, since alot of the light hitting the glass winds up back in the aquarium anyway. Ditto if you have the thick side of the eggcrate down.

 

On the other hand, if the aquarium were filled with water, *essentially all the light hitting the glass would escape*. In that case, the downward-focusing effect of the eggcrate might actually be significant. (Whether the gain is more than what you lose from the opacity of the eggcrate is another question. Based on these measurements). The website http://www.phy.ntnu.edu.tw/ntnujava/index.php?topic=205.0 has a good visualization of this. n2 is the index of refraction of the glass (1.5), n1=1 for air, n1=1.33 for water.

 

It would be interesting to redo these measurements with the tank filled with water to see how that affected the amount of light escaping. You'd probably want the meter to be parallel to the glass faces.

 

Based on what I can see in my tank, which might be a fluke, nooks and crannies and corners are better illuminated with the eggcrate. Overall, the illumination of the tank seems to be less, but I'm basing this on looking at the vertical face of live rock which is towards the back of the tank, which might not benefit very much from downward-directed light. We also tend to gauge brightness by looking at the center of the tank, the corners might actually give a better indication of what's going on.

 

Having said all this, my guess is that you're loosing about 10% of your overall light with eggcrate, but nooks and crannies benefit significantly.

[DaveS]

I think the reason your nooks and crannies are better illuminated is because eggcrate is NOT a light focusing device but a light diffusing device. That's generally why they put it on fixtures. If it was meant to be a focusing device, it would be linearly parabolic- like the way T5 reflectors are. As it stands, what eggcreate does is take a linear "point source" and changes it into multiple point sources. Each grid surface is, by reflection/diffusion, a small source of light. As result, you get illumination over a great surface. The drawback is that each surface has less intensity as seen by the empirical results.

[Brian Ward]

I think the reason your nooks and crannies are better illuminated is because eggcrate is NOT a light focusing device but a light diffusing device. That's generally why they put it on fixtures. If it was meant to be a focusing device, it would be linearly parabolic- like the way T5 reflectors are. As it stands, what eggcreate does is take a linear "point source" and changes it into multiple point sources. Each grid surface is, by reflection/diffusion, a small source of light. As result, you get illumination over a great surface. The drawback is that each surface has less intensity as seen by the empirical results.

 

Dave,

I don't remember if you were there when Anthony was discussing this, but his statement was to use the light diffusers in the reverse position so they would take the lightbeams that were spreading through the glass (since MH reflectors tend to induce spread) and focus them straight down into the tank.