Minimalistic multichip DIY LED build

A little HELP. I made the LED jump 2 months ago after reading this thread, and my LPS are doing GREAT! BUT my SPS are loosing COLOR (they look DULL) and some are dying off. A little about my system.

Tank Dimension: 6 feet (length) x 3 feet (width) x 2 feet (height)

Original Lighting System: (10 inches above water line)
250 Watt 14000K Metal Halide Bulb - Phoenix, Double-Ended (FOUR) - 5 hours
ATI 60 Inch 80W Blue Plus T5HO Fluorescent Bulb (TWO) - 8 hours
ATI 60 Inch 80W Purple Plus T5HO Fluorescent Bulb (ONE) - 8 hours
ATI 60 Inch 80W Aquablue Special T5HO Fluorescent Bulb (ONE) - 8 hours

NEW LED SYSTEM DIY: (16 inches above water line)
50W Actinic Royal Blue Hybrid Led (FOUR)
- 24 chips 10000K + 24 chips 445-460nm Royal Blue
- Seller = led-hk
- Runs at 40% intensity
- No optics
- Ramps up and down, total 5 hours

10W 455nm Royal Blue High Power LED (TEN)
- 452-455NM
- Runs at 100% intensity
- No optics
- Ramps up and down, total 8 hours
- Seller = luoxiquan

10W White 15000K 20000K High Power LED (TEN)
- 15000K-20000K
- Runs at 30% intensity
- No optics
- Ramps up and down, total 3 hours
- Seller = luoxiquan

I don't have PAR meter, but if I run my 50W LEDs higher than 40% things start to bleach out!!
What am I missing?
- Do I need LEDs in the 420 to 430 wavelength?
- Do I need to change my intensity? Should I go further down on 50W LEDs?
- Do I need decrease or increase my total lighting time?

ANY help will be appreciated. Thanks again to Lassef and others who originally answered my startup DIY questions!
 
Lassef said:
I do not know but I had some problem with a "dirty" 1-10 V signal from my GHL Profilux 3 ex. It was some small, but fast variations in the signal and my LED flickered. It was not a meanwell driver but one of type "Steves LED" A friend of mine solve that through adding a type of filter that took the average during some short time and send that to the driver.

Sincerely Lasse
Click to expand...

What he probably added was a capicator. The capiocator you could say stores a small amount of power at the voltage it is connected to. If the voltage were to drop the capicator would slowly discharge trying to hold the voltage at its prior level till it ran out of steam. Simularly if the voltage is increased it would suck up the excess untill it reached equiliberium with the source. These are used often in non switching power supplies to remove the ripple from the DC power. There draw back is they prevent instant on instant off situations.
 
@ Bowfront90

The dull color on your sps probably caused of to low intensity of the white chip´s. My experiences with this type of high Kelvin multrichip says that a ratio of 1:1 or even sometimes 2:1 (white:blue) doing great according to color rendering. With color rendering I mean how you how you perceive the reflective colors of your corals. Can you try the opposite, ie 30% intensity of your 10 watt RB and 100% intensity of your white chip. Just to see if you get a better color rendering - keep the 50 watts at the intensity you have now or mayby a little bit lower. The blue wavelenghts don´t looks bright for the human eye - but they have a lot of energy

Now - corals has often other type of colors than reflective colors. For some reason, a phenomenon called fluorescence is common in many corals. It means in brief that the organism receives a photon of a certain wavelength (energy and color) and brings back a photon of a different wavelength (usually longer, ie less energetic and with a different color). The wavelengths which are often involved here is around 420 nm. We call it "color pop-up". So it may be wise to incorporate some sources with 420 nm wavelength. 420 nm is not the only one involved wavelength but often the most common.

@ TropTrea: Thanks

Sincerely Lasse
 
EJN said:
Gunlocators, I am in the exact same boat as you. My tank is about the same size and I need to replace bulbs asap but I think that I want to switch to LED instead of purchasing new bulbs. I just finished reading through this entire thread (took a week) and have also been considering JPCCUSA's build as well. He is using 48 Cree LEDs. I really like the multichip idea but due to the size and expenses I am leaning towards the 48 Cree rather due to cost and how well his design is layed out. I think i can just follow the plans exactly as he did and have a great light.

Does anyone have any ideas (pros/cons) about the two designs? I would like to compare the two and get the best design for my tank and budget.

Thanks for the great thread and tons of information. Between this thread and the other one I have learned so much about LEDs.[/QUOTE

Have you tried PMing JC to get his input on his fixture to date. ie, how his corals are doing, any changes he would make on a do over, ect. He should be able to give you some good info.
Click to expand...
 
Ok an update... Work has kept me very busy, I have had a baby and I bought a new house so it has been an interesting few months...

BUT...

I have finally have all components I need to complete my build.

The new LED lineup looks something like this

3x 10000K 50w (90* lens)
5x 453nm-455nm 10w (90* lens)
5x 445nm-447nm 10w (90* lens)
1x 430nm 10w (no lens)
2x 420nm 10w (no lens)
2x 405nm 10w (no lens)

Drivers for the 3x 50w

http://www.ebay.com.au/itm/221058281428?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649

And 20w

http://www.ebay.com.au/itm/221110419441?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1497.l2649

I bought 20w chips but i am going to run them at 10w for now.

I was looking at this driver

http://www.ebay.com.au/itm/85-265VA...986?pt=LH_DefaultDomain_0&hash=item3371803a1a

but it is dimmed with a 20k pot and my controller uses a 10k signal (non PWM) and i did not want to play around with but I may look at them later if I want to go with 20w.

Anyway let me know what you think.
 
bowfront90 said:
A little HELP. I made the LED jump 2 months ago after reading this thread, and my LPS are doing GREAT! BUT my SPS are loosing COLOR (they look DULL) and some are dying off. A little about my system.

Tank Dimension: 6 feet (length) x 3 feet (width) x 2 feet (height)

Original Lighting System: (10 inches above water line)
250 Watt 14000K Metal Halide Bulb - Phoenix, Double-Ended (FOUR) - 5 hours
ATI 60 Inch 80W Blue Plus T5HO Fluorescent Bulb (TWO) - 8 hours
ATI 60 Inch 80W Purple Plus T5HO Fluorescent Bulb (ONE) - 8 hours
ATI 60 Inch 80W Aquablue Special T5HO Fluorescent Bulb (ONE) - 8 hours

NEW LED SYSTEM DIY: (16 inches above water line)
50W Actinic Royal Blue Hybrid Led (FOUR)
- 24 chips 10000K + 24 chips 445-460nm Royal Blue
- Seller = led-hk
- Runs at 40% intensity
- No optics
- Ramps up and down, total 5 hours

10W 455nm Royal Blue High Power LED (TEN)
- 452-455NM
- Runs at 100% intensity
- No optics
- Ramps up and down, total 8 hours
- Seller = luoxiquan

10W White 15000K 20000K High Power LED (TEN)
- 15000K-20000K
- Runs at 30% intensity
- No optics
- Ramps up and down, total 3 hours
- Seller = luoxiquan

I don't have PAR meter, but if I run my 50W LEDs higher than 40% things start to bleach out!!
What am I missing?
- Do I need LEDs in the 420 to 430 wavelength?
- Do I need to change my intensity? Should I go further down on 50W LEDs?
- Do I need decrease or increase my total lighting time?

ANY help will be appreciated. Thanks again to Lassef and others who originally answered my startup DIY questions!
Click to expand...

the symthums your describing are typical of coral bleaching from excessive red light especialy in the 580 nm range. Since we do not have spectro graphs of all your chiips it is hard to oin ooint acuractly which is doing it. However I could bet that one of white LED's is very rich in red light.

As far as the 420 nm LED's go this another one where a little is good for a tank and most corals but in excess it can be just as bad if not even worst than red light. If yuou add them by conservative with a max of about 10 watts total for your color balance.

I would try reducing you intensitry slighly on your white kights. one set at a time. While neither sound rich in red that does not mean something got through production and they are producing more red than normal. Bhe other culperates could be the Royal Blue ans Royal blues can be either pure 450 nm chips. od a combination of 465 nm blue and 680 nm Red to give the same visual purple tint to the naked eye.
 
BeanAnimal said:
Dan,

I peltier is a sorely inneficient device and the operational cost would easily negate any advantage that the LED has with regard to efficacy compared to MH or T5.

To be clear, an IDEAL thermoelectric junction is maybe 10% as efficient as the IDEAL Carnot device (phase change refrigerator). So a real world refrigerator is maybe 40% efficient and a real world peltier is maybe 5% efficient. As Lasse mention, not only is it not efficient, its byproduct (it uses energy) is MORE heat to the room.
Click to expand...

I agree, they are compact, but certainly not efficient. Peltier devices are good far a small Killifish bowl or making a frozen food dispenser, but that's about it.
 
007Bond said:
Have you tried PMing JC to get his input on his fixture to date. ie, how his corals are doing, any changes he would make on a do over, ect. He should be able to give you some good info.
Click to expand...

Great idea. I'll PM him and ask if I am allowed to pm otherwise I'll ask him in his thread. Thanks.
 
Lassef said:
I have received some questions about how things are going with the "5 channels multichip". Right now it's about 30 interested people from all over the world who want to be on this journey into the more or less unknown. The technical solutions are largely done and the whole thing will become a reality in the very near future. So far it has been an amazing experience to work with different people through forums, memos and e-mails to have this to happen.

What is revolutionary about this chip is that it comes with a small footprint and that it provide five independent channels, where both intensity and current can be regulated individually. This technique can be used in completely different applications by replacing the LED we have in the various channels.

We have chosen that to channel 1 have 20 white LEDs at 10 000K, on Channel 2 have 20 RB 445 nm, on channel 3 have 10 Actinic 420 nm + 10 Actinic 430 nm, on Channel 4, we have 20 RB 455 nm, and the channel 5 additional 20 pieces white at 16 000K

The reason for the choice is two: growth and to be able to watch. The blue is selected considering various "chlorophyll's" absorption peaks. The white to provide additional wavelengths have other biological roles than chlorophyll, and that there should be an okay color rendition. The low intensity of the yellow-green region is chosen to disfavour organisms that use these wavelengths for photosynthesis.

You will be able to control the light the way you want it - during periods when no one looks at the aquarium, use a light combination that promotes growth and in times someone looks at the aquarium you choose a light combination that provides an aesthetic value.


If this test falls out well, for it is still in the pilot stage, I can see a great potenial in other areas where it is of great importance with differentiated light at different times, it's just changing which LEDs to use, ie, using LEDs with other wavelength regions.

Once everything is set, and different people have started to build their applications, we will of course write about this and learn from each other's experiences and skills. If it will be in a thread here on RC or not I do not know. I understand that RC has some policies that are easy to break in such a thread. This is because a build thread about this has to be closely followed by links to other websites and opinions of different products that are identified by name and manufacturer.

This means perhaps that these discussions will take place in a forum that has other rules. This is a pity because I think (for some reason :)) that we might hold the replacement for metal halide in our hands. At least, in terms of application in biological environments.

Sincerely Lasse
Click to expand...

This is the kind of solicitation post RC doesn't like.
 
troptrea said:
a good rule of thumb to start with is 5 1/2 square inches of surface area on the heat sink per actual watt led's without any cooling fans. By adding vooling fans tou can go to about 3 1/2 square inches of cooling surface. Some of the new designs with liquid cooling however allow you to go much smaller.
Click to expand...

thanks
 
bowfront90 said:
A little HELP. I made the LED jump 2 months ago after reading this thread, and my LPS are doing GREAT! BUT my SPS are loosing COLOR (they look DULL) and some are dying off. A little about my system.

Tank Dimension: 6 feet (length) x 3 feet (width) x 2 feet (height)

Original Lighting System: (10 inches above water line)
250 Watt 14000K Metal Halide Bulb - Phoenix, Double-Ended (FOUR) - 5 hours
ATI 60 Inch 80W Blue Plus T5HO Fluorescent Bulb (TWO) - 8 hours
ATI 60 Inch 80W Purple Plus T5HO Fluorescent Bulb (ONE) - 8 hours
ATI 60 Inch 80W Aquablue Special T5HO Fluorescent Bulb (ONE) - 8 hours

NEW LED SYSTEM DIY: (16 inches above water line)
50W Actinic Royal Blue Hybrid Led (FOUR)
- 24 chips 10000K + 24 chips 445-460nm Royal Blue
- Seller = led-hk
- Runs at 40% intensity
- No optics
- Ramps up and down, total 5 hours

10W 455nm Royal Blue High Power LED (TEN)
- 452-455NM
- Runs at 100% intensity
- No optics
- Ramps up and down, total 8 hours
- Seller = luoxiquan

10W White 15000K 20000K High Power LED (TEN)
- 15000K-20000K
- Runs at 30% intensity
- No optics
- Ramps up and down, total 3 hours
- Seller = luoxiquan

I don't have PAR meter, but if I run my 50W LEDs higher than 40% things start to bleach out!!
What am I missing?
- Do I need LEDs in the 420 to 430 wavelength?
- Do I need to change my intensity? Should I go further down on 50W LEDs?
- Do I need decrease or increase my total lighting time?

ANY help will be appreciated. Thanks again to Lassef and others who originally answered my startup DIY questions!
Click to expand...

It's hard to give advice on whether you have too much light, not enough, or the wrong kind of light without at least PAR values.

First off, your photoperiod should be 8-10 hrs and you are running between 3-8 hrs. Secondly, with no optics on any of these chips, you will not get adequate intensity for SPS. I use 100w multichips minimum for tanks that are 24" or taller. You are running an array of 10w multichips without specifying how close they are together. I don't have any experience with this kid of hybrid system, but in my opinion it gives you the worst of both worlds, the poor spectral distribution and shimmer of an array with the heat issues of a multichip.

Your generic chips may also have strange spectral qualities, as they rarely reflect the colour temperature numbers they are assigned by the manufacturer. I agree, a broader spectrum of blue light would help, but if I was to wager, I would say that you aren't getting enough light. Your new LED system is 400w and the outgoing MH was 1320. LED is more efficient than MH & T5, but you will likely need another 200w with smaller wattage multichips.

Perhaps someone out there has experience with these chips and can be of more help, but I haven't seen any colourful SPS tanks using them yet. I offer my apologies in advance to the 20 people who are about to post pictures of their colourful 10w multichip tanks :)

- add lenses
- increase photoperiod
- test PAR and raise or lower light accordingly
- possibly add more 50w or 100w chips
 
TropTrea said:
As far as the 420 nm LED's go this another one where a little is good for a tank and most corals but in excess it can be just as bad if not even worst than red light. If yuou add them by conservative with a max of about 10 watts total for your color balance.
Click to expand...

He claims that his tank looked great with 160w 420nm T5, so 10w max seems low to me. The only reason I can see limiting 420nm LEDs is the added cost and lack of brightness according to our human limitations. This is more of a mass production issue (making money) than DIY concern.
 
You must log in or register to reply here.

Similar threads

adova
300g DIY Acrylic Tank Recommendations
Replies
5
Views
339
griss
griss
Brian60077
Building another new tank
2 3
Replies
49
Views
1K
Brian60077
Brian60077
C
Maxspect jump led issue
Replies
0
Views
77
Cruzersmith
C
E
Neptune DDR diy question for AWC
Replies
3
Views
360
Reefing102
Reefing102
A
How to Install Linear Actuators for Easy Lifting of Your Reef Tank SPS Light
Replies
4
Views
337
oreo57
oreo57
Back
Top