Energy absorbed by water illuminated by an LED source

[Apgomes3]

Good morning,

I was wondering if I am thinking correctly. I am trying to establishing some step by step to calculate the amount of energy is being absorbed by a water body when a LED pendant is shining right above.

Being the parameter that I want to calculate, the energy absorbed by the water ( to calculate a proper heat exchanger to cool it down), what would you recommend as the best approach ?

I have the LED lux (I can get other parameters as well - as I have the light study for this specific light) more or less around 19103lx @ 1m

A bit lost in how to tackle properly.

Regards

 

[berkeman]

Can you say more about the application and the characteristics of the tank? The energy absorbed by the bottom and walls of the tank will likely dominate, and then the heat conduction from those to the water versus the outside environment need to be factored in.

Is this for an aquarium like in your previous thread from early this year? Are you worried that having such a strong illumination will heat the aquarium a bit too much for the fish inside, and that's why you are wanting to figure out how much heat you need to remove from the water? What are the dimensions of the tank?

In case it is for an aquarium application, I'll page @BillTre

 

[Apgomes3]

Hi Berkman, yes it will be geared towards the aquarium industry. The earlier thread I ended up doing all the calculation and ended up actually being good summary for all heat transfers; i could compute that into a calculator which was my initial goal so i am pretty happy with that (while understanding each step)

what you said is correct, the heating of the tank needs to be managed. what I am failing is to understsand what part of the light is being absorbed by water in energy forms; what part is being absorved by the tanks walls ( I understand that the materials will have different coefficients) so mostly trying to find my way to compute a clean way to assume this. Water plays here as refraction. I also assume that spectrum will play a role in what is being absorbed or not (this is a blue shifted spectrum typically - so I would assume the tanks walls are doing most of the absortion as you said); but then again what can i assume as being reflected energy ?

There is also another one that is the heat dissipation that is occurring in the layer of air above the water tank, but that one is easier to find a way to calculate.The intent is to have a design a calculator to input any tank size ( so it spits out areas, side walls etc) - I have typically used concrete, FRP and acrylic so I account for the different materials properties, but we could for a exercise example do a 3000 x 1000 x 500 mm tank in FRP (white colour) which is tha tank I am looking at now.

Typically the aquarium industry doesnt worry much about this topic, but since this is a shallow tray and being used for culturing corals the amount of light above is considerable (3600 Watt of light)

 

[anorlunda]

what I am failing is to understsand what part of the light is being absorbed by water in energy forms; what part is being absorved by the tanks walls

What difference does that make? If the walls are warmed by the light, the heat in the walls will then warm the water via contact.

A warm wall will also heat the surrounding air by contact. But the wall-to-water heat transfer coefficient should be much higher than the wall-to-air coefficient.

 

[Tom.G]

Water seems to absorb most wavelengths other than visible and near IR. That said, probably most absorption will be by the coral itself.

To further your research you would do well to get the spectrum of the lights you are using. What is the actual power usage of the LEDs? You mentioned 3600W, is that the actual LED power consumption (unlikely, that would be roughly 100Amps), or is it the incandescent equivalent of brightness?

See: http://hyperphysics.phy-astr.gsu.edu/hbase/Chemical/watabs.html

(above, and many more, found with: https://www.google.com/search?&q=optical+absorption+of+water+by+wavelength)

Cheers,
Tom

 

[Apgomes3]

Hi @Tom.G greatly appreciated your input.

See below the reference for spectrum and energy of light pendant in use. This is a rack of 4 lights being tested out. so around 1200 Watt. for the area mentioned above it will be around 3 racks (3600 watt)

Following your commment in terms of light we consider that water is not absorbing (at least in such shallow tanks) any visible light and the animals (corals) are. The issue is that the energy the corals are absorbing is significantly less than the one being delivered; moving away from that a bit so I need to

account for some coefficient for reflective radiation on the water surface (not sure how relevant it is)
The remaining part of photons will then be going two routes; reflected back (depending again in the material and color) of the tank walls or alternatively absorbed by them.

Are you seeing a different iteration to my line of thought?

The reason I am doing this iterations is that I have both - and HVAC to deal with air temperature, and a heat exchanger in a cooling system to deal with water; but want to properly estimate the loads in each.

 

[Tom.G]

Even though the data sheet is rather blurry, it did help.

The data sheet shows optical power in Lumens, which is based on the color response of human vision. The commonly used reference color is 555nm, Green, which is about in the middle of the visual spectrum.

At that 555nm wavelength 1lumen = 1/680lightwatt, or one lightwatt = 680 lumens. Therefore the 100,000 lumen panel would be 147lightwatts if it was all at 555nm.

To get an accurate calculation you would have to account for the human color response and the spectral output of the light panel. The Blue peak in the panel spectrum may complicate this calculation, due to the unreadable graph labelling I can't tell if that is still visible.

Anyhow, that about exhausts my capability on this subject; hopefully there are others here that can pick up this thread for a more detailed answer.

Oh, for air conditioning you can assume that all of the electrical input is converted to heat the air, that is very slightly conservative but much smaller than the normal safety factor for AC. (filters clog, dirt build-up on heat exchangers, slow leakage of refrigerant, the 'hot side' heat exchanger of the AC may be in direct Sun, etc.)

Since this seems to be a long-term project, you might consider redundant systems, especially for environmental control. The building I'm in just had an air conditioning failure of one unit after only 6 years; the expected delivery of the repair part is 4 to 6 weeks away.

Cheers,
Tom

p.s. Please keep us updated, this sounds like an interesting project. Are any details available on what the project is about?

 

[Apgomes3]

Hi Tom,

Thank you very much for the reply. Greatly appreciated it.

If time permits I will structure the calculator in such a way that is shareable as open source; so it can be dissecated and commented out by the forum.

I am sorry, I didnt confirm that the image would be so blurry. later i will share a more crisp version; probably need to share as pdf if allowed.

I work in the Aquarium/Aquaculture industry. This project in particular is for a small coral farm being built. The major long term idea behind this is to slowly create a tool that helps detailing a bit more the thermodynamics; some of the projects I get myself involved tend to be on the larger volume side (>100 m3), and these areas are typically assumed as safety values; a MEP engineer will come and assume for example all heat dissipated to air in a LED light, which is OK when he is sizing the building HVAC together with the Life Support System Chillers together, but many times these are separated.

So on a longer term is mostly me wanting to go into structuring in all the details...

 

[Apgomes3]

Forgot .. definitely the blue peak is in the visible side; when speaking circadian rythms, corals etc. .. the peak at 440 nm is very common in the marine aquarium industry; the term "blue shifted" spectrum is actually common I would say. But green drives more energy so it gets mixed a bit as well. I will continue to share as I get some iterations, or assumptions.

 

[berkeman]

The building I'm in just had an air conditioning failure of one unit after only 6 years; the expected delivery of the repair part is 4 to 6 weeks away.

Good thing it's winter! Phew!

 

[hutchphd]

I reiterate the question

What difference does that make? If the walls are warmed by the light, the heat in the walls will then warm the water via contact.

A warm wall will also heat the surrounding air by contact. But the wall-to-water heat transfer coefficient should be much higher than the wall-to-air coefficient.

Most of this discussion seems irrelevant to me.....needlessly detailed.

 

[erobz]

To calculate heat transfer by radiation I would start by calculating the view factor between the bulbs and the surface of the water.

Here is paper that has calculated some common geometries:

http://webserver.dmt.upm.es/~isidoro/tc3/Radiation View factors.pdf