The Planted Tank Forum banner

Modified Viparspectra 165w Build

1 reading
14K views 54 replies 7 participants last post by  kjacks  
#1 · (Edited)
Hello. I have a 180 gallon planted tank and 3 Viparspectra 165w lights over them. While They look ok, they are more for reef lights and more in the blue spectrum. Growing plants especially reds have trouble. I have modified the beam angle by replacing the lens to 120 degree so they aren't as focused. Currently running the lights on channel 1 (white) at 50% and channel 2 (blues) at 3%. Still not happy with the color, I've decided to modify the leds themselves. I don't have much experience with this but it doesn't look that hard to do.

Playing around with Spectra, I've come up with this spectrum and layout of leds, trying to mimic some spectrum profiles of better planted tank lights I've seen. Also trying to keep the original white channels, and modify the blue channel, keeping some voilet 420nm, uv410nm. Sunrise and sunset doesn't really matter as these channels will both come on at the same time. I couldn't find 12000K leds on Spectra so I used 8000K instead. I would like to get some thoughts on this as I have no idea what these charts on spectra mean. Also, not even sure how they will look visually once modified and mounted over the tank. Am I even on the right path? Any suggestions will be greatly appreciated! Heres the original layout and how I plan to modify it.

Original:
Image

Modified:
Image


Spectra charts:

Image
Image
Image
Image
 
#2 ·
I have nothing for you by way of technical expertise. As someone who is going to have one of these lights in the near future, I am very interested in how you are going to modify it though.
 
#3 ·
As soon as you add whites or have whites best to shoot for the highest CRI you can get..
Score changes with peak heights, not really worth a lot in general.. Guidelines at best.

1.7 Royal blue and 16 cool white (8000k) 17.7 12000k whites.. ;)
or 24 at 10600K ( average of 12000 and 6500k whites)

Since you aren't going to change the white channel saved it as a Spectra file and imported:

*
MIXING LIST
----------------------------------------
myData viparwhite.txt [120°] x1
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 3,096 lm
Radiant flux : 10,328 mW
PPF : 45.6 umol/s
TCP : 9260 K
CRI : 80
λp : 453 nm
Color : #C5C8FF
----------------------------------------

* PERFORMANCE @ 45cm
----------------------------------------
Irradiance : 5.4 W/m²/s
Illuminance : 1,622 lx
PPFD : 23.9 umol/m²/s
----------------------------------------
Now I can add 27 more diodes..
Keep in mind there is practically an infinite number of ways to do this..well a big number.. ;)
I'll keep what I can.

New channel:

I'll get back to you..
Do you like pink over teal?
 
#4 ·
I was hoping you would chime in @jeffkrol 😀. And that’s genius to get the 12k color I didn’t know you can do decimal points. I had a feeling the 100 score wasn’t really necessary as many high end freshwater lights don’t seem to be in that optimal range shown. Im not sure on the color pink or teal, leaning towards teal but I’m ready to be en-lightened. Haha. If this was your light how would you optimize it only adjusting the 2nd channel?
 
#5 · (Edited)
viparchannel

SUGGEST just save it, scan it not open/save.

CSV of channel 1 you can import it into Spectra.
Import/find it/save. Use generic diodes
You can build ch2 around it

I really had a hard time trying to balance the ch1 into anything w/ high cri though wasn't using dimming
I know you did it just fine .. Different thought process.

Sooo decided to go a more "out of the box" look..
You are familiar with the RGB concept of white and the punch it can give to colors.
RGBAmberCyan is arguably better.
sooo
I abandoned CRI in favor of gamut (exaggeration) Red squares are the orig diodes.
Tried warm whites to balance the cool whites in the other channel. Wasn't working well.



Image

Image


It looks best w/ the orig channel dimmed at around 50%
CRI never goes above about 80-ish...
So feel free to accept/reject/change to your hearts contentment

I ran my estimate with your ch2. Minor discrepancy..
* MIXING LIST
----------------------------------------
myData viparwhite.txt [120°] x1
LED UV (410nm) [120°] x2
LED Violet (420nm) [120°] x2
LED Green (520nm) [120°] x8
LED DeepRed (660nm) [120°] x6
LED NeutralWhite (4000K) [120°] x10
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 5,051 lm
Radiant flux : 18,440 mW
PPF : 82.2 umol/s
TCP : 7140 K
CRI : 92
λp : 453 nm
Color : #EFE6FF
Oddly at 50% on the orig channel the 2 are quite different panels are relatively similar.
Like I said, I sacrificed "natural" for a bit of punch.

Comparison..Ignore power output..just a generic guess.
All calcs are based on inputted lumens and beam angle.
ADA RGB..
* MIXING LIST
----------------------------------------
myData adargb.txt [120°] x1
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 1,000 lm
Radiant flux : 3,305 mW
PPF : 14.5 umol/s
TCP : 11110 K
CRI : 83
λp : 453 nm
Color : #82E9FF
----------------------------------------

* PERFORMANCE @ 45cm
----------------------------------------
Irradiance : 1.7 W/m²/s
Illuminance : 523 lx
PPFD : 7.6 umol/m²/s
----------------------------------------
 
#6 · (Edited)
Awesome thanks for the detailed response! Honestly have no idea what I’m doing just looked things up today and slapped something together. It’s great to see your thought process in making this.

Just some questions. How do I import the csv in to spectra? In The data tab there’s supposed to be a csv tab but it isn’t there.
How do I lower the intensity? I rarely run above 50% as these lights are insanely powered. Also the leds on the spectra site are 1w. Mine would be 3w. Does that make a difference?

I see that your spectrum improves on the amber and cyans I like it 👍. Why is it 8 590 nm or 4 590nm, 4 660nm? Also I see this overall color profile is cyan. Is the other choice for pink overall? Any pictures of pink vs cyan lighting in a tank and what are the advantages of each?
 
#7 · (Edited)
Awesome thanks for the detailed response! Honestly have no idea what I’m doing just looked things up today and slapped something together. It’s great to see your thought process in making this.

Just some questions. How do I import the csv in to spectra?

Click DATA:FILE:Navigate to the CSV file:click on it, preview should pop up. You can change beam angle and lumens here: SAVE


In The data tab there’s supposed to be a csv tab but it isn’t there.
CSV button is for export...not import

How do I lower the intensity? I rarely run above 50% as these lights are insanely powered. Also the leds on the spectra site are 1w. Mine would be 3w. Does that make a difference?
The 1w and 3 watt thing is just if you using the wattage/par calculations.
If you import your own data (long story involving multiple softwares and manufactures spectrums and electrical characteristics)

Well all this is a long story .. to make it a bit short the light you have and the diodes you will probably buy are run at about 2 watts regardless of their "class"
designation.
A "3 Watt class" diode run at 550mA and a V(f) at that amperage of 3.5 volts is actually 2 WATTS.
Really using 1W diodes is probably a fair approximation of your light running at 50% on each channel


I see that your spectrum improves on the amber and cyans I like it 👍. Why is it 8 590 nm or 4 590nm, 4 660nm? Also I see this overall color profile is cyan. Is the other choice for pink overall? Any pictures of pink vs cyan lighting in a tank and what are the advantages of each?
the all 590 vs 4 590 4 660 is just to punch up reds more and 660 are photosynthetically more efficient. Well 660 hits chlorophylls a bit harder than amber.
Less wasted photons so to speak.
Image


Difference is sort of like the 2 red sq photos..
Image


To "fix" it to "magenta tone" overall you would need to remove the 470's and replace them with like regular red..
* MIXING LIST
----------------------------------------
LED UV (410nm) [120°] x2
LED Violet (420nm) [120°] x2
LED Cyan (490nm) [120°] x5
LED Green (520nm) [120°] x6
LED Amber (590nm) [120°] x4
LED Red (630nm) [120°] x4
LED DeepRed (660nm) [120°] x4
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 1,293 lm
Radiant flux : 6,138 mW
PPF : 27.7 umol/s
TCP : 5080 K
CRI : 82
λp : 413 nm
Color : #FFAB9A
I can't get you best of both worlds.. i.e dimming one channel pushes you across the cyan/magenta divide.

Hmmm. the cyan and orig white channel end up fairly similar.
Dimming one or the other doesn't spectrally change much. May be different in person.

2new channels
 
#8 · (Edited)
Thanks for all this valuable information I appreciate it. 👍 I like the look of the cyan more than pink/magenta but do you think plants will look too green/washed out compared to the magenta? What color are you leaning towards in your tanks?

I’ll plug it into the program and go from there. Already ordered the leds and I’ll post updates here as well so people can see the progress.
 
#9 · (Edited)
Hmmm I think my orig file was wrong.
This would be for the channel you won't touch.
The orig white channel
It was fine in my SPECTRA.Anyways reposting
Original White Channel

Think my preference is with the magenta one.

Take the 2 files:
viparorig.csv and viparmagenta.csv and "dim" orig by multiplying magenta.
Changed beam angle and checked th box to force it to use 90 degrees.
Ch1 dimmed 50% New channel (magenta) 100%


* MIXING LIST
----------------------------------------
myData viparmagenta.csv x1
myData viparorig.csv x0.5
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 2,840 lm
Radiant flux : 11,299 mW
PPF : 50.5 umol/s
TCP : 6940 K
CRI : 93
λp : 663 nm
Color : #FFD1E9
----------------------------------------

* PERFORMANCE @ 45cm & 90° (compulsory)
----------------------------------------
Irradiance : 17.8 W/m²/s
Illuminance : 4,465 lx
PPFD : 79.4 umol/m²/s
I'm a bit uncomfortable with the violets. (420,410) The calculator shows a heavy influence on the color.

Need one more calculation..
Need to estimate the V(f) total of the modified board with the added reds.
Since you are going to be opening the box check the driver . It should have something like
Output 60-90V DC
550mA
Quick estimate is 73-79.8 Volts total for the 27 diodes.
40-43 Watts for that channel

If the driver voltage minimum is higher than 73 Volts there might be an issue.

One picture shows 70-90V, current 550mA so that year driver would be fine. :)
 
#11 ·
What did led selection did you choose with the original channel? So to "dim" we essentially just change the beam angle? I can remove the violets if its not needed. Now to plug this all in. However I realized the upload file button is hidden by scanning the html. I think this is due to my computer being a mac. 🤦‍♂️ Now to boot up my ancient dell computer.
 
#12 · (Edited)
2 -660
2 -520 green
6 -6500 k (used 8000k)
18 -12000k (used 16- 8000k + 1.4- 440 royal blue)

You can dim by using fractions
Or if you aren' t using the power data just add in proportion.

Say ch 1 is "1" and ch 2 is "2" is the same as
Ch 1 at "0.5" and ch 2 at "1".

Beam angles are used to calculate Photons per meter squared per sec.
 
#13 ·
@jeffkrol Just want to pick your brain for a bit. What are your thoughts on 2hraquarist website which shows ideal spectrum profile of ada solar as the following, mostly only rgb?

Image


And then me playing around with channel 2 with white channel at 50% and getting something like this both both channels combined:

channel 1: is as your settings.

channel 2:
1 410nm
2 420nm
3 450nm
16 520nm
5 630nm



Image
 
#14 ·
If you want to go with that tone I'd suggest 660 for 630 and 470 for 450..
* MIXING LIST
----------------------------------------
myData viparorig.csv [120°] x0.5
LED UV (410nm) [120°] x1
LED Violet (420nm) [120°] x2
LED Blue (470nm) [120°] x3
LED Green (520nm) [120°] x16
LED DeepRed (660nm) [120°] x5
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 3,013 lm
Radiant flux : 10,871 mW
PPF : 48.1 umol/s
TCP : 9430 K
CRI : 93

λp : 663 nm
Color : #80FFED
@jeffkrol Just want to pick your brain for a bit. What are your thoughts on 2hraquarist website which shows ideal spectrum profile of ada solar as the following, mostly only rgb?
You always need to start with "ideal for what"?
His show tank..
Image


There is sort of a err "cultural divide" so to speak. Pretty sure created by marketing.
1) ADA/ Amano ect wit h heavy emphasis on green/cyan and US probably since there was a heavy dose of "grolux" growlights in the magenta/pink range..
same thing w/ saltwater.. Europeans tend to favor "whiter" reef lighting while the US favors more blue reef lighting..

As 2 cri you have natura lvs exaggerated.. high cri white vs rgb.

ADA Solar:
* MIXING LIST
----------------------------------------
myData adargb.txt [120°] x1
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 1,000 lm
Radiant flux : 3,305 mW
PPF : 14.5 umol/s
TCP : 11110 K
CRI : 83
λp : 453 nm
Color : #82E9FF
----------------------------------------

* PERFORMANCE @ 45cm
----------------------------------------
Irradiance : 1.7 W/m²/s
Illuminance : 523 lx
PPFD : 7.6 umol/m²/s
----------------------------------------

Mr. Wongs light is like 4500K-ish

Plants adapt they can use all the spectrum and pretty much equally .

One last diode .. This is a "full spectrum" diode.
Blue pump broad red phosphor(s).
FULLSpectrum
 
#15 ·
Got it. Exactly the thought process I was looking for. So there seems to be a school of thought where people want more spectrum than just rgb. That is a beautiful tank. If I wanted a look like that, which I know it won't be close, keeping my original white channel, would that be the pink or magenta channel 2 you described? Or something entirely different?
 
#16 ·
This is my deep red x 3, cyan, pcamber, lime, royal blue pucks I made.
Apologies for the weird placement.


I can find his pattern. It really is just r,b,white.

Image
 
#18 ·
You can have a magenta rgb. Only difference is balance.
2:2:1 bgr... cyan tint 2:1:2 bgr magenta tint .

B:G:R
B:G:R:Amber
B:G:R:Amber:Cyan
MIXING LIST
----------------------------------------
LED RoyalBlue (450nm) [120°] x1
LED Cyan (500nm) [120°] x3
LED Green (540nm) [120°] x1
LED Amber (590nm) [120°] x3
LED DeepRed (660nm) [120°] x2
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 518 lm
Radiant flux : 1,970 mW
PPF : 9.2 umol/s
TCP : 6540 K
CRI : 97

λp : 663 nm
Color : #FFFCFE
----------------------------------------

* PERFORMANCE @ 30cm
----------------------------------------
Irradiance : 2.3 W/m²/s
Illuminance : 611 lx
PPFD : 10.9 umol/m²/s
----------------------------------------

by SPECTRA 1.0β @ 1.023world
R:G:B:W
R:G:B:CW:WW


Color theory...
Why some white leds are dull
 
#20 · (Edited)
From posted voltages on flea bay vendors ..
They line up with "history" though quality control can be so bad that the same "lot" of blue may have V(f)'s of 3-4V (rough guess)

Bottom line of that paper..
Blue "pumps" or blue phosphors with yellow phosphors is color dulling..

It's a matter of comparing personal "preference" in look with combinations of "RGB" phosphors.
With some other stuff ..;)
It does not account for the "newest" best white leds using either extended phosphors or violet or 2 blue pumps....

You can roughly equate ind. colored leds as a
phosphor color ie cyan phosphor= cyan led.

Then there is how characteristics change with increases in intensity.
And lots of silly abbreviations.
 
#21 · (Edited)
@jeffkrol What do you think about the sbreef freshwater led layout? I plugged in the specs into Spectra and looks like they mostly use a white channel like mine, a lot of reds and blues with cyan mixed in. What surprised me was the color spectrum looks so low in the middle portion. It also uses 55 lights exactly so I could replicate it. Cri seems extremely low though. In pictures of the setup the tank does look darker than others since its low in the green spectrum.


* MIXING LIST
----------------------------------------
myData sbwhite [110°] x1
myData sbcolor [111°] x1
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 3,430 lm
Radiant flux : 22,141 mW
PPF : 101 umol/s
TCP : ‑ K
CRI : 21
λp : 663 nm
Color : #B864FF
----------------------------------------

* PERFORMANCE @ 50cm
----------------------------------------
Irradiance : 13.5 W/m²/s
Illuminance : 2,115 lx
PPFD : 62.1 umol/m²/s
----------------------------------------

by SPECTRA 1.0β @ 1.023world


Image



Image



Image

Image
 

Attachments

#24 ·
It's sort of odd all around..
It's not "missing" anything unless you hang a hat on IR/UV.

With 2 channel designs my "philosophy" if you want to call it that is 1 warm, one cool. Blending the 2 should come out to your target
look..

I'm doing this relatively fast soo..
A
4-6500
4- 660
4-3500
4- 640
4- 490
6-590
2- 640
(28)
* MIXING LIST
----------------------------------------
LED Cyan (490nm) [120°] x4
LED Green (520nm) [120°] x2
LED Amber (590nm) [120°] x6
LED DeepRed (640nm) [120°] x4
LED DeepRed (660nm) [120°] x4
LED CoolWhite (8000K) [120°] x4
LED WarmWhite (3000K) [120°] x4
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 1,898 lm
Radiant flux : 7,011 mW
PPF : 34 umol/s
TCP : 3640 K
CRI : 81
λp : 643 nm
Color : #FF8D61
--------------------------------------
B...
Like this:
Averaged the 6500/10000k for expediency. Need to subtract one from each or 2 from the 6500k
* MIXING LIST
----------------------------------------
LED Blue (460nm) [120°] x4
LED Green (530nm) [120°] x3
LED DeepRed (660nm) [120°] x6
LED CoolWhite (8000K) [120°] x10
LED WarmWhite (3000K) [120°] x4
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 2,180 lm
Radiant flux : 8,625 mW
PPF : 39.6 umol/s
TCP : 9710 K
CRI : 62
λp : 457 nm
Color : #DFBDFF
----------------------------------------
* MIXING LIST
----------------------------------------
LED Blue (460nm) [120°] x4
LED Cyan (490nm) [120°] x4
LED Green (530nm) [120°] x3
LED Amber (590nm) [120°] x6
LED DeepRed (640nm) [120°] x6
LED DeepRed (660nm) [120°] x10
LED CoolWhite (8000K) [120°] x14
LED WarmWhite (3000K) [120°] x8
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 4,016 lm
Radiant flux : 16,035 mW
PPF : 76 umol/s
TCP : 6250 K
CRI : 85
λp : 662 nm
Color : #FF97A6
----------------------------------------

* PERFORMANCE @ 45cm
final answer..
You can generate the 2 channels to verify..

I left out any violets but you can tweak the "cool" channel a bit..
I'd suggest only 410's..
Replace 2 660's B channel

Hmm got a bit turned around..
B: w violet:Remember 8000k is 6500/10000 combined )
300 bit lower than the 3500k
* MIXING LIST
----------------------------------------
LED UV (410nm) [120°] x2
LED Blue (460nm) [120°] x4
LED Green (520nm) [120°] x3
LED DeepRed (660nm) [120°] x4
LED CoolWhite (8000K) [120°] x10
LED WarmWhite (3000K) [120°] x4
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 2,125 lm
Radiant flux : 8,812 mW
PPF : 38.8 umol/s
TCP : 12660 K
CRI : 52
λp : 457 nm
Color : #BFACFF
----------------------------------------

* PERFORMANCE @ 45cm
----------------------------------------
Irradiance : 4.6 W/m²/s
Illuminance : 1,113 lx
PPFD : 20.3 umol/m²/s
----------------------------------------
Final sort of.. You can check and clean.. Only a ballpark estimate..
W/ 120 lenses all around. If a deep tank switch to 90's or if shallow raise w/ 90's..
Also remember this is w/ 1W diodes.
myData sbreefA.csv [120°] x1
myData sbreefB.csv [120°] x1
----------------------------------------

* SIMULATION DATA
----------------------------------------
Luminous flux : 4,022 lm
Radiant flux : 15,817 mW
PPF : 72.8 umol/s
TCP : 6670 K
CRI : 95
λp : 662 nm
Color : #FFAFC4
----------------------------------------

* PERFORMANCE @ 45cm
----------------------------------------
Irradiance : 8.3 W/m²/s
Illuminance : 2,107 lx
PPFD : 38.1 umol/m²/s
----------------------------------------
Image


Ch a, b and "c" (carotenoids)
Image
 
#26 ·
Question. How does one replace the diodes on their gro light and automate the dimming? I have a mars hydro sp3000 which only comes with a manual dimmer. I used to use Arduino based controllers when I had separate cobs but I like the profile of these modern gro lights more and would love to mod them.

From what I can see viparspectra is similar and doesn't come with it by default.
 
#28 · (Edited)
As long as they are still using the Bridgelux" egg" diodes removing and replacing is relatively easy.
There are reports of some boards using NORMAL smd diodes. THAT is a lot more difficult to deal with.
Anyways normally the removal is fairly destructive to the diode.
And some are soldered in the center, some just use thermal compound.
Oh make sure the new ones hav thermal compound and I suggest the "glue type. Thermal silicone like "heatsink plaster"
No need for the more expensive 2 part thermal epoxies. Though they say hard to remove, I've never found that but
why take the chance. The thermal silicone adhesives are easier to deal with down the road.
I don't suggest just plain thermal compound since the diodes can lift easier nor thermal tape.
A teeny bit works fine on the center metal circle. You'll see what I mean.

As to controlling.. THAT gets a bit more complicated but fairly easy.
Depends on the model.

Determining the orig drivers dimming protocol is the first step.
Sometimes it's labeled on the "pot" board.

One complication is it usually also includes an "off" circuit since the dimming doesn't "go to zero"..
There are a ton of writeups mostly on reef forums.

Short story is you need to convert the 5V (usual) PWM of the Aduino to the required signal for the driver.
Usually 10v analog (1-10v) or 10V pwm for the manual dim models. If one wants to bypass the models with on board programming well completely different story.

Jumper one wire to bypass on/off switch and use a timer on the AC plug to shut it off completely at the end of your cycle.
Oh and don't program below like 10% dimming. Drivers can get flakey.

Actually that is sort of adjustable if you crack open the driver itself BUT generally unnecessary..

This isn't the only way.. There are add on boards that are pre-built.


4 wires..see you need to know what you have.
Writeup mostly for apex which outputs 10v analog. Using pi aduino ect you need to still convert its signal to 10v
 
#30 ·
I've been kinda following along though a lot of the jargon is above me. While you guys were going back and forth I did contact SBreef lights and asked for the freshwater board. I know this was mentioned already but I figure I will throw in that its 75$ shipped. I am ordering it to replace the original board in the viparspectra board in the soon to by mine used light I am getting.

Here is the layout picture they gave me.

Image
 
#32 ·
I've been kinda following along though a lot of the jargon is above me. While you guys were going back and forth I did contact SBreef lights and asked for the freshwater board. I know this was mentioned already but I figure I will throw in that its 75$ shipped. I am ordering it to replace the original board in the viparspectra board in the soon to by mine used light I am getting.

Here is the layout picture they gave me.

Image
Post some pics when you get it!

Leds are here. Here we go wish me luck. First couple were tough to get off it involves desoldering then chiseling the old leds off with a micro screw driver and several taps with a hammer. I think I’m getting a hang of it though just took off two without breaking them. I wouldn’t recommend doing this if you’re not extremely patient and like diy projects. First one probably took 30 mins to get off.



Image
 
#33 ·
Bit concern you are using the hammer to put them in ;)
-
One question why do this instead of buying a pre-built light like wrgb-2 or sbreef or whatever. Is it for the 'fun'; is it cheaper; is it that you are looking for a specific look or whatever that pre-configured products lack; is it .... ???
 
#36 ·
Leds are here. Here we go wish me luck. First couple were tough to get off it involves desoldering then chiseling the old leds off with a micro screw driver and several taps with a hammer. I think I’m getting a hang of it though just took off two without breaking them. I wouldn’t recommend doing this if you’re not extremely patient and like diy projects. First one probably took 30 mins to get off.



View attachment 1040479
Well ..most just twist them off. Of course most just break.
At about 50 cents each, not an issue really.
At one point the center wasn' t even soldered so it was just thermal compound.

Anyways still fun...

Go to about 1O min..

So it is cheaper to buy viparspectra and then modify them? I'm mostly asking this question since i'm trying to decide how i will light the new tanks i'm buying (8ftx4ft and 10ftx3ft).
Cheaper than what?
You got me thinking ..
There are horticulture lights as well
So first a commercial method .
NEVER MIND. Thus was going to be based on sbreef fw replacement boards.atm I don't see those available. Cost was around $70 though.
So upgraded light would be $150-$200
$80 about 110 watt light with lenses.
2 retail ac/dc constant current drivers could run $35 each .Basically box/ fans/diodes/ lenses are err " free".
New diodes are 50 cents or less each..
Most spectrum adjustments to make it 100% or so useable for fw ...say 20 diode minimum.
$10 using the reef version though there us a premium charge on them.

Sooo add say $130 plus $10 and time and solder/thermal compound
I' ve noticed that $2-$3 per watt is pretty normal for average lights. Rarely $1/watt

I'll get back to the hort version if I can find the diode arrangement.
Just needs green ( to make white).

Sadly not dimmable but 2 seperate channels, reflectors not lenses. 5 w diodes.
200 real watts total.

Sorry off topic a bit..