The Planted Tank Forum banner

One Way to Design a Planted Tank LED Light

82K views 42 replies 20 participants last post by  Hoppy 
#1 ·
Here is one way to design a LED light that will meet your planted tank requirements. Obviously, you first start with the tank you want the light to work with. Next, decide if you want the light to be suspended above the top of the tank, or sit on top of the tank. Suspending it is by far the best idea. Decide if you want to spend a little more money and be able to adjust the apparent color temperature of the light, which requires more LEDs and two LED controllers.

A very good PAR to design for is 60 micromols per square meter per second. This is medium intensity light - if you miss the target, you should still have either low-medium light or high medium light, both of which are very usable.

Pick the LEDs you want to use: Cree XP-G, which are high output, high power, and a bit cheaper than Cree XM-L, which are higher power, and higher output.

If you will suspend the light above the tank, you will want to use optics on the LEDs to focus the light to reduce the spillover light, and make better use of the full output of the LEDs. For lights which will be less than 2 feet from the substrate, 60 degree optics work well. For lights up to at least 36 inches from the substrate 40 degree optics work well. Decide which you will use.

It is best to run LEDs at less than maximum power to ensure that they will last for many years before “wearing out”. Also, the more power you run them at, the more likely you will need active cooling and a quality heatsink, which add to the cost, the complexity, and the noise. Pick a LED driver or drivers which can deliver the current you want to use.

Look up the lumen output the LEDs produce at that current from the pdf data sheets for the particular model LED you will use.

Write down these parameters:
LED model _______________
LED current_______________
lumen output per LED_______ (L)
Height of light from substrate________inches (H)
Optic cone angle__________(probably 60 or 40 degrees) (ø)

Now make this calculation:


where “a” equals the spacing between LEDs in inches, and “n” equals the number of rows of LEDs to be used.

Assume one row of LEDs will be used, and calculate the spacing between LEDs ( the square root of the result above)

Alternatively, assume 2 rows of LEDs will be used, and again calculate the spacing between the LEDs.

Assume 3 rows of LEDs will be used, and again calculate the spacing between the LEDs

To decide how many rows of LEDs will be best, calculate the spread of the high intensity portion of the cone of light. That will be equal to 2 x H x tan ø/4. That spread should be about 2/3 to 3/4 of the front to back depth of the tank. If it is less than that, 2 or 3 rows of LEDs should be used, to expand the spread.

Now that you know the spacing of the LEDs, and the number of rows of LEDs, calculate the number of LEDs per row. The row of LEDs can end at about one half of the tank front to back depth from each end of the tank, to avoid spilling a lot of light at the ends of the tank. The length of each row will be the tank length minus the tank front to back depth. Divide that number by the spacing of the LEDs (a) to get the number of LEDs per row.

Look at the specifications for the LED drivers you want to use to see the minimum and maximum number of LEDs they will power. Based on that you can determine how many LED drivers you need. If you want to be able to adjust the apparent color temperature of the light you will have to use two colors of white, cool white and neutral white, for example, with separate dimmable drivers for each color. Make sure the number of LEDs of each color is not fewer than the minimum for the driver.

Electrically connect each color of LEDs all in series, up to the maximum the LED driver will power, and connect those to one dimmable driver. Do the same for the other color.

Pick a heatsink or heatsinks that will accommodate the rows and LEDs per row that you will use. If you are using an LED current near the maximum power for the LED model you are using, also figure out a cooling fan or fans configuration to keep the heatsinks cool.

Finally, determine how you will hang or otherwise support the light, and find the parts needed to do that.

Use of a “splash guard” is optional. Lights hanging above the tank do not need any splash guard. Lights sitting on top of the tank, with no glass lid on the tank, may need a splash guard, and will need one if you use an air stone in the tank.
 
See less See more
1
This post has been deleted
#6 ·
Here is an excel spreadsheet that embodies Hoppy's formula to calculate the PAR (top portion) and the spacing between the LEDs (lower portion). I included Hoppy's suggestion to round off the PAR value. The spreadsheet is protected except for the input cells to prevent loss of the formulas, but you can unprotect it if you are so inclined (there is no password).

View attachment 44811

I notice that your spreadsheet also requests to use "half the optic cone" angle. Which is a reference to other posts where Hoppy and other have said that this is the area of most intensity.

You do use the same equations Hoppy used above though. IE, you're still dividing the cone angle by 4.

Is this how the equation in the original post of this thread is meant to be used? It makes sense to use the "effective angle" for PAR (half the optic angle, so 20 for a 40, and 40 for a 60), but making sure that Hoppy did not already account for this in his equation?
 
#8 · (Edited)
The angle is the advertised cone angle of the optic, not half of it. I corrected this on the spreadsheet, and added a calculation for the LED current for Cree XM-L and Cree XP-G LEDs. But, I don't know how to attach an Excel spreadsheet to this post. Once I know how to do that we can include both the correction and the added calculation.

I used the Cree pdf for each LED model and derived an equation that works for most of the range of the Lumens vs current graphs here. The equations work for the 350 to 1500 mAmp range for both of them. For the XP-G the equation breaks down for lower currents, and for the XM-L it breaks down for higher currents, overestimating the lumens produced, in both cases. But, I wouldn't even consider using a higher current for either LED, just because of the cooling that would then be required for the heatsinks.

Thanks to fizzout I learned how to to this: Here is the updated calculator, for which fizzout gets 95% of the credit.

Edit: Added Cree XP-E-Q5 LEDs which are available at DealExtreme, and which are more suitable for lower height tanks.
View attachment 44862

Edit: Added Cree XR-E-P4 Bin WD (5700-6350K), which are less than $3 each on DealExtreme. View attachment 44874
 

Attachments

#10 ·
And with that the light design philosophy for the next 1X years has been written.

Think I'm pretty well set that I want to do an LED next then. Will pick up some valuable soldering/electronics skills along the way.

I know I read Hoppy in another post that the "number of rows" multiplication may not hold up once the spacing between rows grows to the 6" distance which would happen if I spaced out even 3 over the middle 12 of an 18" front-to-back, so I will have to use this thread, other graphs Hoppy's posted, and probably a little guesstimating to figure out how to build a 40B light. I will need to crunch some numbers in Excel if I get a chance this week but thinking I'll try to do it with three rows, but treating the PAR value as "two" rows because of the wide spacing between the outermost front and back. I am presuming, before having done the numbers, if I aim for 50-60 PAR with 600-700 mA this three row design would get me close enough that I could either go down intensity or up, if I'm wrong, and not hit the "min current" or "need cooling" thresholds.

Use of LEDs will allow me to build a light to fit my desired canopy design, instead of building my canopy around desired lights (for T5HO I'd need to be concerned with heat as well as height, on a 40B I will want a "squatter" canopy)



Sent from my BlackBerry 9650 using Tapatalk
 
#15 ·
Hoppy,

I did some work using the equations and posts you’ve made in this thread and elsewhere along with the excellent spreadsheet tool. I think my questions below belong in this thread because they address a topic that will be common among anyone with wider tanks.

I am working on designing a fixture for a 40 breeder which is 18” front to back. As far as I can find online I haven’t found anyone making one yet so breaking some new ground, in a way.

I don't mean the below to ask, "design me a fixture." Instead I am asking for some clarifications or suggested assumptions in using the equations and sheets to account for cases of widely spread LED rows of 4-5" spacing between rows (and >8" spacing between first and third rows)

As previously suggested by yourself and common sense, I went into this with some “knowns” and requirements for the fixture, to fix some of the variables. I know that I want the recommended middle PAR range of 40-60 as you have frequently referenced, I want even lighting over my tank, no more than 2 LED drivers, dimmable for fine-tuning PAR, passive cooling via aluminum channel or other heat sink, and I wanted to build a canopy that is less than 15" above the tank rim - the canopy height bit is one of the more important variables that can be fixed because it most directly impacts how the LED will be designed of course.

For use in the spreadsheet I chose 20” above substrate as the height. Pending actual builds etc I could vary this up and down within reason, as the canopy itself won’t be built prior to the lights.

I know that an 18” wide 40 breeder with lights 20” above the substrate will require multiple rows of LED lights if I use aluminum channel passive cooling. With 40 degree optics (more on this in a second), at a height of 20”, the “high intensity” inner-half cone diameter will have a 3.5” radius at the substrate surface. To avoid high intensity glass strike, I figured that the starting point for my rows would be to have one directly centered in the tank, one 5 inches from the front glass and one 5 inches from the back glass.

I read with great interest a post you made elsewhere regarding that rows of LEDs greater than a few inches apart don’t add up when kept at a low height, and thus it probably isn’t accurate to use “n = 3” for number of LED rows in this design.

So, knowing that my front and back rows would be 8” apart, and the center row is 4” from the row on either side of it, it seemed I was at the point where it’s just a guess to select any number above 1 as the “number of rows” combining for PAR.

Using 1.25 as “n” rows to account for the large spread from front row to back row but still account for “some” combination, I came up with the following:

27 Cree XPGs
3.333” Spacing per row, 4-5 inches between center row, front row and back row
20" above substrate
60 PAR
40 Degree optics

Would let me run the XPG at .6-.7 Amps – within the passive cooling range and if I adjust the PAR down it gets better from there.

This seems like a high number of XPGs, and with quite focused Optics, compared to what I’ve seen elsewhere used for tank lighting applications. Do you think I have made drastically wrong assumptions at some point in the process?

If I change to 60 degree optics the calculated LED current required to hit 60PAR with the 27 XPG (3 rows of 9 at 3.33”) goes to 1.1 A. Way too high for passive cooling. But why would I need 40 degree optics only 20 inches above the substrate?


edit: Of course if I change my assumption to "n=2" which is still of course only 66% of my actual number of rows to account for the spread from front to back, it decreases the number of LEDs required (4" spacing) and amperage as well. Seems like I might need to just experiment. Which isn't so bad anyway - the two worst cases are that I either end up having to go higher above my tank than I wanted to do and possibly rethinking using a canopy, and/or having an extra row of LEDs on a heat sink I can use for another tank. I can't find it but I am sure you have posted somewhere the graph of Cree XPG output vs distance from light (horizontal) at one or various mA? If you have, I would be able to look at that and figure out how much each spot on substrate should get with this setup.
 
#16 ·
Doc7, you ask a number of good questions:biggrin: The equation that this method is based on was derived from data that I have from LED lights with rows of LEDs spaced about the same distance as the LED spacing. But, the maximum LED spacing I used was 3.25 inches max., and the number of rows was 3 maximum. Extrapolating from that will work, but probably not for 3X extrapolations. The higher above the substrate the light is, the more you can extrapolate, just because the cones of light get bigger as you go higher. (I think it is about the overlap of the light cones, more than anything else.)

I suspect that with 3 rows of LEDs, 20 inches from the substrate, and 60 degree optics, and 4-5 inch row spacing, you can use n=3. At 20 inches the center half of the cones of light is about 10 inches in diameter, so there is a lot of overlap of the rows of light. I'm not sure that with 40 degree optics you can use n=3, and I suspect you can't. But, for a 20 inch height I wouldn't use 40 degree optics anyway. A 40B tank is about 17 inches high, so, with 3 inches of substrate, the top of the tank is about 14 inches from the substrate, and a light 20 inches away is about 6 inches from the top of the tank. A 40 degree optic would give a circle of light about 4 inches in diameter at the water surface. These circles of light are very visible when they are that small, and wouldn't look good at all.

The PAR numbers you get with those spreadsheet calculation is for right under the middle of the light, so the distance between the outer rows isn't relevant. As with any aquarium light the PAR will drop off near the glass anyway - theoretically. But, light reflected off of the glass adds to the PAR near the glass, greatly reducing the drop off.
 
#17 ·
All white or a mix?

This is some fantastic information. I have been trying to determine the correct number and type becuase I want to move away from t-5's. As I am reading this it seems to implicate a fixture made up of all white LED's. I have had many recommend a mix of cool white and blu or royal blue. Some at a 1 to 1 ratio and others at a 2 to 1 ratio. What are your thoughts on this? And does the calculation still work for the white only when doing a mix?

Thanks!
Joe

38Gal 30x12x19 planted tank
56Gallon 30x18x24 soon to be planted
95 gallon 36x24x25 soon to be planted discus.
 
#18 ·
Question on optics angles

Hoppy, thanks a lot for the work. I've been planning a LED build for my 200 gallon, full spectrum, and have pieced together a spreadsheet with spectral output of various LED's and what I'd need to have to get xxx illumination. I get numbers fairly close to yours for numbers of LED's.

I've got a question on the spreadsheet (ver 5), related to the half angle/full angle question stated earlier. The documentation says that if you don't use a lens for the LED's, that

Cone angles for bare LEDs, with no optics installed, XP-G, XM-L, XP-E use 70 degrees, XR-E use 50 degrees
When I look at the spec sheets those look like half of the full width half maximum (FWHM) for the bare LED's (for example, for the Cree XPG the spec sheet the spatial distribution is 50% of maximum at +/- 60 degrees, therefore 120 degree half maximum or FWHM).

So should the angles for the bare emitters be 140 (or 120) and 100? Or am I miscalculating what a "60 degree optic" is by assuming it's +/- 30 degrees at FWHM?

Thanks.
 
#24 ·
The calculator relates six variables. You provide five of them and it calculates the sixth.

You need to answer the following question: what do you want it to calculate ?

Then you need to provide data for all the other five variables. From your statement you already have:
LED height = 48
Cone angle = 30

If you can get the value of three more variables, the calculator will give you the value of the sixth.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top