Lassef
Member
That works also but my suggestion was one driver to 3 chip - the chip in parallel mode
Sincerely Lasse
Minimalistic multichip DIY LED build
- Thread starter maglofster
- Start date
That works also but my suggestion was one driver to 3 chip - the chip in parallel mode
Sincerely Lasse
ReefUrchin
New member
That's what i was talking about, the not killing you thing
Well, if I go with lasse's configuration, I will be running 7 separate channels at 1.0amp min on one channel. I want them all dimmable but I really dont want to buy 7 PS's and would like to condense. I mean at $100 each, thats $700? Another reason I was looking for the most universal spectrum blue LEDs, to try to run only two channels. Running them all together is going to bring up the Amperage to? Whats the 12v to amp danger point?
This is gonna be a tricky build!
ReefUrchin
New member
That works also but my suggestion was one driver to 3 chip - the chip in parallel mode
Sincerely Lasse
How is the dimming? Same characteristics or? Any other drawbacks/undesired effects going parallel besides the obvious covered?
Last edited:
Spotlight?
Spotlight?
Quick question for the experts... My tank is a 110 Tall, 48" X 18" X 30", and I have two 100 watt chips and a number of 3 watt chips to supplement them. I am going to be designing my fixture here soon and was wondering what you guys would recommend for spacing?
Would it be better to build a more compact fixture for a single light source type look? Similar to running a single high wattage halide...
Or should I spread the fixture over a larger foot print to get better spread?
That being said my limiting factor is that I only have about 12" above the water to place the fixture... I have a full machine shop and can make anything so I was even thinking about making the chip mounting angle adjustable if that would help?
Any help would be greatly appreciated!
Thanks,
Jeremy
Spotlight?
Quick question for the experts... My tank is a 110 Tall, 48" X 18" X 30", and I have two 100 watt chips and a number of 3 watt chips to supplement them. I am going to be designing my fixture here soon and was wondering what you guys would recommend for spacing?
Would it be better to build a more compact fixture for a single light source type look? Similar to running a single high wattage halide...
Or should I spread the fixture over a larger foot print to get better spread?
That being said my limiting factor is that I only have about 12" above the water to place the fixture... I have a full machine shop and can make anything so I was even thinking about making the chip mounting angle adjustable if that would help?
Any help would be greatly appreciated!
Thanks,
Jeremy
Lassef
Member
Both of these post concerns if it grows better when concentrate light into wavelength ranges from 420 to 460 nm. I think so - at least, you need to put less energy to get the same results. It is fact that the photosynthesis of most corals occurs in these wavelengths. Potentially also at 470 nm. The red wavelengths are usually less important because they disapear with increasing depth. Furthermore the longer wavelengths does not penetrating so deep in turbid water over a reef either.
Other wavelengths are certainly important too but when associated with additional pigment that converts incorrect wavelengths (read energy quanta) to such a level that the photosynthetic active substances can handle it. This usually costs energy so if you can aim for the interesting areas directly it is better. Technically, it has been difficult in the past but now the technology have been developed so that LEDs in these wavelengths are relatively inexpensive (relatively).
But there is drawbacks with this. Here is a translated post I did in the swedish reef forum on the same issue.
UV usually counts below 400 nm (UV-A) in the normal context. 420-430 is violet also called actinic. RB is a floating term first used by Cree and became very famous. Then there came out a lot "RB" in the market. What distinguishes AC-RC from other right now is that he puts out the wavelengths (or K white) - not just only "trade names". Initial RB - Cree - wavelength range you can see below.
RB is far left. You see that the peak is around 460 nm and at 420 nm, only about 25% of the intensity that it gives at 460 still left - eg photons is about 75% less. The area that I think is most important is 420-455, so you should concentrate on these wavelengths. What I wanted to say is that to get as much light at 420 nm compared with Cree RB it’s enough with 1/4 as many watts (Watts is a bad parameter but unfortunately the most useful right now) This means that one must be careful about switching from Cree RB to these wavelengths below 460 nm. You cannot use the experience of Cree's RB to determine the strength of this construction. If you also use 455, 445, 430 and 420, you have to keep in mind that the intensity at 420 nm is much higher than just its own intensity, as the other three also contribute. And of course, vice versa if you look to the intensity at 455 nm. This is the cause why this configuration was chosen for the Dream chip: 20 pcs 455 pcs 20 445, 10 430 and 10 420 If I should take up the wavelength I think is most important as it 445 nm. It and the 430 are the ones that get maximum total intensity of dream chip design.
But back to being cautious when starting the usage of these wavelengths. I want to emphasize once again - be careful - start at a low dose - increase slowly over a few months.
If you take a look at this graph
Datasheets are taken from Cree because I do not have access to the spectra of the white from RC right now - it will.
Here you can see a peak at about 455 (which means that it is the blue LED that is covered with phosphor to get the white light) but if you see the surface of the charts so is the most quantity of light at other wavelengths. It is stronger blue peak higher K. Now different phosphor mixtures from each other so no direct transfer to AC-RC's high Kelvin chip cannot be done, but the blue is less prevalent there.
Context is a bit that follows:
As it is, all experience how much blue light you need for a good growth is based on Cree RB. Cree RB has its optimum at 460 nm - at 420 nm, the intensity is only about 25%. At 445 about 40%. With the dream chip or other construction including the wavelength range of 420 – 460 nm, we go directly to the wavelength peaks we aim to. Its mean if we put 1 watt at 420 nm - we get 1 watt. If we use Cree RB, we get only 25% of the intensity at 460 nm. To this must be added the contributions to 420 nm from 430, 445 and 455 nm. The construction of the Dream chip is based on maximum power around 430 -445 nm. It will be very interesting to see the spectra further on.
But once again - be very carefully with the introduction of this chip or all other construction including the wavelength range of 420 – 460 nm. If we maximize photosynthesis too fast for our corals it goes bad. They need time to develop protection against the increased production of free oxygen radicals
I hope I'm completely clarified my position: Be careful out there
Sincerely Lasse
ReefUrchin
New member
Makes perfect sense. It will just take a lot of math, LED layout and some good old logic to duplicate the "RB" stats and adjust from there for my particular setup. I am going to have to sit down and look at some fixtures and see what is working, how many watts are being run yada yada yada, lol.
Thanks Lasse
miatawnt2b
New member
Anyone have experience with the 30000k chips?
TropTrea
New member
This is realy going to be determines by the effect you are tryiong to create. If you go to two point sources and all the led grouped together as tightly as possible in thise two points you will get some strong shadows and fantastic shimmering effects as the light reflects on the surface.
If you spread the lights out equaly over the entire surface area then you have a much more even light distribution with less noticable sharp shadows. However you will also loose most of the shimmering effect.
When I lighted up my 120 gallon I used 60 3 Watt LED's evenly spaced throughot the tank but the front row was angles at a 45 degrres to the rear and the back row was angled at 30 degrees to the front. I was able to get shimmer and yet even distribution that way. But if I were building it today I would go with 24 10 Watt LED's evenly spaced throughout the top. This would give me a little more shimmer and a little more shadowing effect without getting drastic. I know some people on this group would advise 2 100 units, and other 8 25 watt units. It is lot about personal preference.
firefightered66
New member
OK:headwalls: I have read most all the posts here and am still so confused because alot of this I do NOT understand. I have a 75 gal 48x18x20ish reef tank. All I want to do is light this baby up to where the corals will be happy and still pleaseing to the eye. what will I need? I do not have a controller so I will need to dime the light mannually with pots or somthing and put them on timmers for now. I need:
What multichip LED and wattage
What driver(s)
Lens
Heatsinks
Ect?
Just want to have a good light source and get rid of my t-5 and MH setup
What multichip LED and wattage
What driver(s)
Lens
Heatsinks
Ect?
Just want to have a good light source and get rid of my t-5 and MH setup
For diffusing the higher power LEDs, has anyone tried an old fashion plastic diffuser like they use to place into the fluorescent fixtures? I am referring to the clear diamond pattern flat sheet. I was just wondering if this might help to deal with the concentrated lights and get better coverage with less tendency to create shadows.
Dennis
Dennis
TropTrea
New member
My recomendation would be a total of about 175 Watts of LED's. Since the tank is only about 20 " deep I would go with ultra high powered multi chips but rather something like 36 5 watt chips, or 18 10 Watt chips. The reason I would go with more chips of lower power is to get a more even light distribution.
A nice combination I would recomment would be 6 CREE XM-L Neutral White LED that are run at 3000 ma and 3.35 Volts for 10 Watts each 12 CREE XP-E Blues running at 1000 ma and 3.5 Volts for 3.5 Watts each, and 16 Cree XP-E Royal Blues running at 1000 ma and 3.5 Volts for 3.5 Watts each.
You can do this on three seperate meanwell drivers.
Whites on a HNL 80-30 wich actualy will run them at 2.7 Amps for 9 Watts each Then the Blues and Royal blues on a two HLN 60H-54 which will run 14 chips each at 1150 ma or 4 Watts each.
For heat sinlks I would use 3 1" X 2" channel Alumnium 48" long. you should not need to have a any fans with this set up as long as the top is open for normal air circulation.
One thing to note about using arrays with different colored emitters (IE not single hybrid multichips) is that if you are running a lid, you can get strange lensing effects from the condensation drops. This is probably not a factor for lower latitutes where the winters aren't so dry but up here everyone runs lids, otherwise the evaporation is just too much, and there can be other bad effects from the evaporation even if you put up with it.
For example, I get spots where the neutral white light is lensed to make an ugly yellow spot, or perhaps the blue is lensed away? Either way it makes it look like a sick spot in my tank. This happens also on my friend's 250 with it's 200x3w emitter array, and it uses 20000k whites. The "ocean coral" emitters are the worst though, they make hideous turquoise and red and windex spots.
This is one of the reasons why I initially started researching multichips. Even with a tight cluster of LEDs I still get disco lights (though it does mitigate some) and odd shadowing.
EDIT: i wanted to add that yes, you can use fluoro style diffuser to help with softening shadows and reducing disco. Some distance between the diffuser and light is needed. It will cut down on intensity and penetration, to the tune of 15-20%. Acrylic is like 93% transmittance but there is also like 7+% of reflectance, depending on angle of light to the surface (dead on is best of course). This bears out in my own tank, PAR with the lid on (no condensation) is about 20-25% lower than with it off. I may run open top eventually (it's only a 12) but not until I setup the ATO.
For example, I get spots where the neutral white light is lensed to make an ugly yellow spot, or perhaps the blue is lensed away? Either way it makes it look like a sick spot in my tank. This happens also on my friend's 250 with it's 200x3w emitter array, and it uses 20000k whites. The "ocean coral" emitters are the worst though, they make hideous turquoise and red and windex spots.
This is one of the reasons why I initially started researching multichips. Even with a tight cluster of LEDs I still get disco lights (though it does mitigate some) and odd shadowing.
EDIT: i wanted to add that yes, you can use fluoro style diffuser to help with softening shadows and reducing disco. Some distance between the diffuser and light is needed. It will cut down on intensity and penetration, to the tune of 15-20%. Acrylic is like 93% transmittance but there is also like 7+% of reflectance, depending on angle of light to the surface (dead on is best of course). This bears out in my own tank, PAR with the lid on (no condensation) is about 20-25% lower than with it off. I may run open top eventually (it's only a 12) but not until I setup the ATO.
Last edited:
Lassef
Member
I have not any experiences with chips as high K as that - is it anyone out there that has try this type of chips?
For some strange reason, I disagree with you
My experience is that the large multi-chip does not create that much shadows in a normal aquarium. The reason is the scattering angle, which is usually about 120 degrees. Do you use lenses, which work differently from MH's reflectors; you get a more even light. Where you can get a little less light is the first 10 - 15 cm unless the sources light cones merges with another already in the surface.However, I have not before been fond of the big chips, at least not in my tank at home. The reason is that I want to have control of the light and to have my light temperature changing throughout the day. I have mostly worked with 10 and 20-watt chip.
The 5-channel multi-chip has changed the landscape for me. Now I can get the best of both worlds
Here, I agree
You're the only one, beside myself, that I have heard using angled light in the front of the aquarium. I have a background that is common here in Europe, especially among freshwater aquarists - BTN - Back to Nature. I have different modules where I could hide my technical equipment. I of course want corals that grow on these "stone" areas. A light from above will not access properly and does not give the right intensity. Therefore, I have a rig at the front facing the back of the aquarium with adjustable angle.Thats true
Sincerely Lasse
Last edited:
Lassef
Member
Interestingly, here in Sweden with a similar climate as Canada - dry indoors during the winter - it's not common with lid. This is in many cases just to get the indoor humidity. Excess humidity is vented out but it is also common with heat exchange during ventilation. I know some have had problems with moisture, but many are also using three glass windows therefore moisture on windows is not so common. I have this problem though
Sincerely Lasse
firefightered66
New member
Thanks for the reply. I was wanting to go with the multichip version for $$$ reasons. Mayby the 10 watt.. What colors and configuration would you suggest?
Landsailor
New member
Heatsink Question:
Shockingly, nowhere in Las Vegas has 2x1 aluminum channel in stock. SpeedyMetals could get me four pieces of 36" 2"x1"x1/8" c channel for about $57 shipped.
However, http://www.rapidled.com/regular-heatsinks-1/ has 1.25" black heatsinks for $25 (shipped) per 36" piece.
That would be $43 more for actual heatsinks with more surface area than the c-channel.
Good idea or wasted money?
Shockingly, nowhere in Las Vegas has 2x1 aluminum channel in stock. SpeedyMetals could get me four pieces of 36" 2"x1"x1/8" c channel for about $57 shipped.
However, http://www.rapidled.com/regular-heatsinks-1/ has 1.25" black heatsinks for $25 (shipped) per 36" piece.
That would be $43 more for actual heatsinks with more surface area than the c-channel.
Good idea or wasted money?
Sonny n Colleen
New member
There is heatsinkUSA and if you want more choices there is this place, http://www.micforg.co.jp/en/index.html
