Moving LED Array to Simulate Tracking of the Sun

herring_fish · 01/30/2013, 05:10 PM

I am doing some preliminary study on designing and building a moving LED lighting array that simulates the movement of the sun. I heard that there is a company that is using this idea to grow some of their corals. Does anyone know what the name of the company is?

I saw one product that is call a Sunspot but I doubt that it would do the job that I want it to do. What I want may not yet exist in this fast changing market.

I want to use a single unit that has powerful LEDs in it that would take the place of more than one metal halide. This unit would be supplemented with a set of blue LEDs that run the length of the tank in a more traditional fashion. Despite the blue LEDs, I might want to tune the color balance to get the best look for me but that may be too much to ask for without jacking up the price. I may instead want to add some other color to the bank of blue LEDs and use them to do the job.

My primary motivation is cosmetic. I just think that it would be much more interesting to look at with the constant changing as the sun spot moves across the tank during the prime viewing hours of the tank.

Since I would only use one lighting unit that passes over the coral, any one coral would get less light than if it had a stationary light over it all day long. So how much brighter would it need to be for my 24 inch deep 130 G? ...the eqivalent of 1,000 watt MH?
equivalent

This is my old unit that I built in the early 90's. I build a 6 inch square polycarbonate tube with a fan at one end. This allowed me to place the lamps only a few inches away from the top of the tank. The air from the fan was forced over the bulbs and cooled the enclosure enough that it would not melt the plastic. The tube channeled the warm air away from the hood and allowed the tank to stay cool. I used mirror polished aluminum reflectors for the lamps. In the graphic, I have partially cut them away so that you can see the bulbs. I used some parts from my sons Erector Set to make a carriage that ran on an aluminum rail on one side and a piece of threaded rod on the other. I drilled and threaded a piece of white UHMW plastic, then cut it in half. I attached that to the blue carriage back plate and it rested on the rod. When the rod was rotated, the carriage moved over the threads to simulate the path of the sun.

I am finishing up a 2 axis robot that I use to feed plankton cultures. I had a steep lerning curve to fight through but now I can use an Arduino mini-controller to control the motor and lighting with comparative ease.

I read some articles about flashing the LEDs on and off very quickly to reduce light saturation which allows users to run the LEDs brighter yet cooler. This also increases photosynthesis and possibly increases the rate of growth while reducing the chances of bleaching. Does anyone know anything about anyone that has implemented this technique at all? Has anyone use an Arduino to do it? I would thing that I would use PWM but I am not an electronics expert so I need help.

There are many fancy low tech things that I could do like getting the lamp the rotate back toward the center of the tank at both ends without additional motors buy using a piece of plastic that is cut to act as a cam or ramp. If I didn't want to waist LED's I could use color theater jells to filter the light from the lamp when it tilts back across that tank to simulate the low angles of the sun and get the changing warm yellows and reds of dawn and dusk.

Instead of using a polycarbonate tube, I am thinking about using some kind of flexible hose with a fan at the other end, to move with the lighting unit and remove heat from the hood.

Can you think of anything else that I should consider or have any information that might be helpful?

mcgyvr · 01/30/2013, 06:45 PM

There are plenty of "light movers" out there.. Pot heads..oops I mean gardeners have been using them for years..

However a light mover is NOT to simulate the sun. Its to reduce the amount of light fixtures required to cover a certain area.
Sunrise/sunset/slope dimming are all possible now with just about all the reef controllers out there. Dimmable LED drivers simply accept a 10V PWM or 1-10V analog or even 100K potentiometer for dimming. Heck some "geeks" are even doing lightning/cloud/storm effects via simple dimming of the LED driver.
Done with the arduino already too.

herring_fish · 01/30/2013, 09:06 PM

Thanks for answering. Yes I am familiar with that. Surprisingly enough, I do remember some of the 70's.

I have concerns that if I move the light, I will either have too much light and get bleaching or too little and get stunted growth or both. I should get more even growth but the amount of light could be an issue.

I hope to get away from a panel because it gives me more freedom. I also have heard about them for years but I have not actually seen storm shows and they sound cool. They are far down on my list for now. I actually want to see the angle of the rays and the shadows change during the evening. I just think that it would be more interesting to me.

I could convert a strobe light for the lightning or have a big array of LEDs to do it. Its just not the direction I am looking to try at first. I also have a design to build the blue array so that each LED can be individually aimed. By the way, I am an over active designer by trade and I am looking for this particular look. I really like to see the rays flashing through the water. I am working toward NPS corals so there are often lots of particulates in the water column which accentuates the rays but now they are fixed and constant. If I was an electronics guy, I would be more interested in storms because they ARE cool.

Now, the "solar flashing" is a feature that may have some particular requirements that I don't know about (beside not knowing how to accomplish it). What is the optimal speed of the flashing? I read in one study that found an on off ratio of 1:10 worked but is that the latest? Of course this can be done with a large flat array or compact ones like I want to build. I need some insight and practical knowledge.

I can do some low level programming and I can cut an paste like nobodies business but electronics is not in my skill set. I just assemble so I need to be led around with simplified instructions so once I get started on the project, I will need some rudimentary instruction.

Thanks again.

herring_fish · 02/07/2013, 01:26 PM

I did some research and found a patent the uses some very fancy method of tuning light output in real time by monitoring the reaction of the subject being lighted. I noticed that they patented this technology to apply the growing many things including corals. Most of the studies that I looked at dealt with liquid algae cultures or leaves but are apparently believed to apply to corals as well. The coral studies accidently came to the same conclusion but were originally set up to look for mitigation of bleaching. This is an excerpt from an algal study but it give you an idea of the general drift of the rest of them.

Quote:

The effect of flashing lights can be theoretically explained using light and dark reactions of photosynthesis. Photosynthesis can be hypothesized as a discontinuous, linear, four-step process [14]. Emerson and Arnold [15] performed the first experiment on the application of flashing light in algal cultures in 1932. When algal cells were illuminated by a succession of very short flashes, it was found that the maximum rates of oxygen production and carbon dioxide uptake under this flashing light could be the same as those under continuous light. This means that photosynthetic cells do not need continuous illumination. For example, the cells will only receive the light for 1/1000 of the time when grown under a 100 Hz flashing light with a flash duration of 10 s, and the cells can undergo photosynthesis as if they had received photons continuously [16]. Other studies were also reported on the effect of various frequencies and duty cycles on the oxygen production rate and the biomass production in algal cultures [17-20].

The existence of dark periods between the short flashes of the light can increase the efficiency of the photosynthesis, especially for a high-intensity light. The aim of this study was to optimize high-density algal cultures by supplying a suitable flashing light, which has higher instantaneous photosynthetic photon flux (PPF) than continuous light with the same average intensity [20]. This study describes a method to estimate the critical cell density and its application to highdensity algal cultures.

I did a little more research and found that having the LED's flash and about 10kHz work well. They should be on about 40% of the time so overdriving or using a higher wattage LED's could be used if desired to get the same average light processing. Most of the studies used peak output compensation to achieve their derived averages.

This offers latitude in light exposure. If a light is to pass over any given coral, more light needs to be used because the time period is reduced. Having higher peak light pulses with rest periods in-between them, allows for more light in a shorter period of time, hopefully without bleaching.

I know that most LED users have a mix of white and blue LED's, sometimes adding other color to round out a good look that appeals to them. most people try to duplicate a look of 14000k to 20000k with some actinics thrown in.

> I am wondering, If I had an abundants of blue lights, would a 6500k spot light look good? ...and would it make for good growth?

zachts · 02/07/2013, 09:22 PM

Should work just the same as your MH version. use a LED unit that provides the nessesary par for a tank of your depth and just move it along so that each 1ft of your tank gets the same photoperiod of overhead light. same concept as the big coral farmers are using. Only they are moving the light from one end of a building to the other over 24hours and then repeating to give 6-8or so hour photoperiod to each small section ofthe setup using only 1 light instead of dozens...........

One could use a single radeon to light a 12ft long tank by moving it allong over the course of a very long day to give every thing a 6-8 hour photoperiod below the light........ Vs using 6 or more radeons for the same thank.......aesthetically that wouldn't work real well and you'd need some supplemental lighting for display purposes to light the whole tank but the heavey lifting could be accomplished by a single moving fixture.

herring_fish · 02/08/2013, 09:12 AM

I heard that one was doing it. you are saying that it is fairly wide spread. Great! Thanks.

Now back to the aesthetic, in the old days, we use to use 6500K metal halides and actinic VHO's. Now, other Kelvin ratings are available for metal arcs so tastes have changed, including mine.

Now, large single LED chips are in their early stages just like the bulbs were years ago. The biggest chips are 6500K but for 12, 14 or 20000K chips, you need to go smaller. I would like to go as big as possible so that I can use flashing to compensate and tune the light output to a higher degree than has been possible before now.

Can a large array of deep blue LED's color correct for the 6500K look?

01/30/2013, 05:10 PM	#1
herring_fish Crazy Designer Join Date: Mar 2008 Location: Winston-Salem NC, USA Posts: 1,029	Moving LED Array to Simulate Tracking of the Sun I am doing some preliminary study on designing and building a moving LED lighting array that simulates the movement of the sun. I heard that there is a company that is using this idea to grow some of their corals. Does anyone know what the name of the company is? I saw one product that is call a Sunspot but I doubt that it would do the job that I want it to do. What I want may not yet exist in this fast changing market. I want to use a single unit that has powerful LEDs in it that would take the place of more than one metal halide. This unit would be supplemented with a set of blue LEDs that run the length of the tank in a more traditional fashion. Despite the blue LEDs, I might want to tune the color balance to get the best look for me but that may be too much to ask for without jacking up the price. I may instead want to add some other color to the bank of blue LEDs and use them to do the job. My primary motivation is cosmetic. I just think that it would be much more interesting to look at with the constant changing as the sun spot moves across the tank during the prime viewing hours of the tank. Since I would only use one lighting unit that passes over the coral, any one coral would get less light than if it had a stationary light over it all day long. So how much brighter would it need to be for my 24 inch deep 130 G? ...the eqivalent of 1,000 watt MH? equivalent This is my old unit that I built in the early 90's. I build a 6 inch square polycarbonate tube with a fan at one end. This allowed me to place the lamps only a few inches away from the top of the tank. The air from the fan was forced over the bulbs and cooled the enclosure enough that it would not melt the plastic. The tube channeled the warm air away from the hood and allowed the tank to stay cool. I used mirror polished aluminum reflectors for the lamps. In the graphic, I have partially cut them away so that you can see the bulbs. I used some parts from my sons Erector Set to make a carriage that ran on an aluminum rail on one side and a piece of threaded rod on the other. I drilled and threaded a piece of white UHMW plastic, then cut it in half. I attached that to the blue carriage back plate and it rested on the rod. When the rod was rotated, the carriage moved over the threads to simulate the path of the sun. I am finishing up a 2 axis robot that I use to feed plankton cultures. I had a steep lerning curve to fight through but now I can use an Arduino mini-controller to control the motor and lighting with comparative ease. I read some articles about flashing the LEDs on and off very quickly to reduce light saturation which allows users to run the LEDs brighter yet cooler. This also increases photosynthesis and possibly increases the rate of growth while reducing the chances of bleaching. Does anyone know anything about anyone that has implemented this technique at all? Has anyone use an Arduino to do it? I would thing that I would use PWM but I am not an electronics expert so I need help. There are many fancy low tech things that I could do like getting the lamp the rotate back toward the center of the tank at both ends without additional motors buy using a piece of plastic that is cut to act as a cam or ramp. If I didn't want to waist LED's I could use color theater jells to filter the light from the lamp when it tilts back across that tank to simulate the low angles of the sun and get the changing warm yellows and reds of dawn and dusk. Instead of using a polycarbonate tube, I am thinking about using some kind of flexible hose with a fan at the other end, to move with the lighting unit and remove heat from the hood. Can you think of anything else that I should consider or have any information that might be helpful?

01/30/2013, 06:45 PM	#2
mcgyvr Registered Member Join Date: Sep 2003 Location: North Carolina Posts: 20,050	There are plenty of "light movers" out there.. Pot heads..oops I mean gardeners have been using them for years.. However a light mover is NOT to simulate the sun. Its to reduce the amount of light fixtures required to cover a certain area. Sunrise/sunset/slope dimming are all possible now with just about all the reef controllers out there. Dimmable LED drivers simply accept a 10V PWM or 1-10V analog or even 100K potentiometer for dimming. Heck some "geeks" are even doing lightning/cloud/storm effects via simple dimming of the LED driver. Done with the arduino already too. __________________ Who me?

01/30/2013, 09:06 PM	#3
herring_fish Crazy Designer Join Date: Mar 2008 Location: Winston-Salem NC, USA Posts: 1,029	Thanks for answering. Yes I am familiar with that. Surprisingly enough, I do remember some of the 70's. I have concerns that if I move the light, I will either have too much light and get bleaching or too little and get stunted growth or both. I should get more even growth but the amount of light could be an issue. I hope to get away from a panel because it gives me more freedom. I also have heard about them for years but I have not actually seen storm shows and they sound cool. They are far down on my list for now. I actually want to see the angle of the rays and the shadows change during the evening. I just think that it would be more interesting to me. I could convert a strobe light for the lightning or have a big array of LEDs to do it. Its just not the direction I am looking to try at first. I also have a design to build the blue array so that each LED can be individually aimed. By the way, I am an over active designer by trade and I am looking for this particular look. I really like to see the rays flashing through the water. I am working toward NPS corals so there are often lots of particulates in the water column which accentuates the rays but now they are fixed and constant. If I was an electronics guy, I would be more interested in storms because they ARE cool. Now, the "solar flashing" is a feature that may have some particular requirements that I don't know about (beside not knowing how to accomplish it). What is the optimal speed of the flashing? I read in one study that found an on off ratio of 1:10 worked but is that the latest? Of course this can be done with a large flat array or compact ones like I want to build. I need some insight and practical knowledge. I can do some low level programming and I can cut an paste like nobodies business but electronics is not in my skill set. I just assemble so I need to be led around with simplified instructions so once I get started on the project, I will need some rudimentary instruction. Thanks again.

02/07/2013, 09:22 PM	#5
zachts Registered Member Join Date: Mar 2005 Location: St. Louis, MO. Posts: 3,259	Should work just the same as your MH version. use a LED unit that provides the nessesary par for a tank of your depth and just move it along so that each 1ft of your tank gets the same photoperiod of overhead light. same concept as the big coral farmers are using. Only they are moving the light from one end of a building to the other over 24hours and then repeating to give 6-8or so hour photoperiod to each small section ofthe setup using only 1 light instead of dozens........... One could use a single radeon to light a 12ft long tank by moving it allong over the course of a very long day to give every thing a 6-8 hour photoperiod below the light........ Vs using 6 or more radeons for the same thank.......aesthetically that wouldn't work real well and you'd need some supplemental lighting for display purposes to light the whole tank but the heavey lifting could be accomplished by a single moving fixture.

02/08/2013, 09:12 AM	#6
herring_fish Crazy Designer Join Date: Mar 2008 Location: Winston-Salem NC, USA Posts: 1,029	I heard that one was doing it. you are saying that it is fairly wide spread. Great! Thanks. Now back to the aesthetic, in the old days, we use to use 6500K metal halides and actinic VHO's. Now, other Kelvin ratings are available for metal arcs so tastes have changed, including mine. Now, large single LED chips are in their early stages just like the bulbs were years ago. The biggest chips are 6500K but for 12, 14 or 20000K chips, you need to go smaller. I would like to go as big as possible so that I can use flashing to compensate and tune the light output to a higher degree than has been possible before now. Can a large array of deep blue LED's color correct for the 6500K look?