High Tide Aquatics

LED Lighting, yet another choice

What about connecting the heat sink to your tank somehow like the old drop in chillers? I'm an idea man and don't know how it would work or how you could control the amount of heat you were putting into the tank, but maybe there's a way?
 
Coral reefer said:
What about connecting the heat sink to your tank somehow like the old drop in chillers? I'm an idea man and don't know how it would work or how you could control the amount of heat you were putting into the tank, but maybe there's a way?

heat exchanger. Jake posted something like this on Reef Builders a while back. Some one was cooling his LED's with the tank water, or rather, using some heat exchanger.
 
Gomer said:
rygh said:
sfsuphysics said:
Not sure the physics of that really make sense.

It takes X energy to light the tank, and Y energy to heat the tank. REGARDLESS OF METHOD.
With LED, X = completely from LED. y = completely from heater.
With MH, x = from MH, y = from MH + heater.

If LED energy = MH energy
and
heater(LED) required energy > heater (HM) energy
because
heater with LED is on longer then MH
because
heat from LED goes to room while heat from MH also goes into tank
then
LED energy + Heater(LED) energy > MH energy + Heater (MH) energy

The problem with that statement : LED energy != MH energy. It is less than half.

Look at it very simplistically:
You get light + heat from both systems.
Which heats the room more (obviously MH by far).
That energy loss is what makes the overall system less efficient.
 
rygh said:
The problem with that statement : LED energy != MH energy. It is less than half.
See that's the problem with LEDs, too many blanket statements that really are not true at all. As it stands the BEST LEDs you can buy right now (not the press released products that Cree is testing) are slightly more efficient (lumens per watt) than the most efficient MH lamps. While its true we don't use those color temperature MH bulbs over our tanks (more yellow end of the spectrum), those who use LEDs toss in those blue LEDs as well which are no where close in terms of efficiency as the white ones, so overall the efficiency from that standpoint drops as well. Then there's the fact that most LED setups you can buy just don't use the best that are out there, at best they use "last generation" best, and at worst, they use the 3rd party knock off to reduce costs on the product. I can't argue with manufacture claims that a 48watt unit = the same output as a 250w halide simply because they probably took the best/worst case scenario... and quite honestly probably fudged the results.

Look at it very simplistically:
You get light + heat from both systems.
Which heats the room more (obviously MH by far).
That energy loss is what makes the overall system less efficient.
Heats the room AND tank more though. Yes I'm counting on some of that inefficiency to heat my tank.
 
This sounds right, heaters should be more efficient because it does not lose energy in Air. & it is on demand better than raising water temp straight up.

But MH warm up room air in winter it is also nice because you save on room heating :)
 
GreshamH said:
PAR readings from LEDs are always suspect to me given the nature of LEDs and the nature of low end PAR meters. They usually read lower then they really are.

Does this mean in reality LED likely output more PAR than it measured?
The PAR measurement is already very impressive on many LED units.
 
As far as overall energy use, going from 250W MH to 120W LED, I realized approximately 130W/ day savings on the electric bill- it's hard to be exact with all the variables of using electricity, but my guesstimated savings were realized.
Regarding the discussion about energy used, heat in tank, etc, the point not emphasized enough is *when* the heat is transmitted to the tank. Heat during the day is a negative, at least in summer for most of us, and heat at night is when it's needed, so the heat of any lighting during day times doesn't equate to energy not used by the heater in the wee hrs in the morning, unless you're willing to have larger temp swings or if you tuck your tank in to bed like Dave does. :D
 
sfsuphysics said:
rygh said:
The problem with that statement : LED energy != MH energy. It is less than half.
See that's the problem with LEDs, too many blanket statements that really are not true at all. As it stands the BEST LEDs you can buy right now (not the press released products that Cree is testing) are slightly more efficient (lumens per watt) than the most efficient MH lamps. While its true we don't use those color temperature MH bulbs over our tanks (more yellow end of the spectrum), those who use LEDs toss in those blue LEDs as well which are no where close in terms of efficiency as the white ones, so overall the efficiency from that standpoint drops as well. Then there's the fact that most LED setups you can buy just don't use the best that are out there, at best they use "last generation" best, and at worst, they use the 3rd party knock off to reduce costs on the product. I can't argue with manufacture claims that a 48watt unit = the same output as a 250w halide simply because they probably took the best/worst case scenario... and quite honestly probably fudged the results.

Hmm. I will try to be careful on my wording on this, since I both agree, and disagree.

Comparing top end MH with junk-LED is not really fair, yet it is true.
Like I mentioned above, and you state, what you get off the shelf for LED is rather weak, and some has some
really cheap Chinese crap LEDs in it. If you see "Prolight", run away. Might be Korean. Whatever.
On the other hand, MH has really been refined through the years, and a nice electronic
ballast and US/Euro/Japanese bulb is really state of the art.

If you are saying MH is theoretically as efficient as modern LED, head-on, sorry, go away. :)
And that specifically means comparing true light energy output versus electricity in.
Even in the not always useful lumens/watt, LEDs are 120-139, where MH are usually 70-85.
Data sheets numbers are pretty easy to measure. Not fudged.

Regarding the color/efficiency argument and blues:
(And in this case - ignoring the junk-leds.)
The problem really is a data sheet issue. LEDs are rated in lumens or sometimes mW.
Rating the blue end in lumens is nearly useless. Since it is a function of the human eye, and we don't see that very well.
Rating in mW is great, but how do you compare it??
So they ARE actually very efficient. Just hard to tell.

---

Overall, the LED "can" do a fantastic job. It is great at converting electricity to light at a given tight frequency.
Way way way better than MH.
*IF* you hit that wavelength correct, it is a huge win.
So if coral was a one-wavelength animal, and you hit it, that 48W -> 250W is probably an understatement.
But clearly coral is not.
SO : It comes down to color!!!!
If you get the right mix of blues, with a bit of red and other spectrums, you win. If not, you lose.
But if that was not bad enough : The human eye color responsiveness is not even remotely the same as coral photosynthetic needs.
Hence the endless problem of mixing and matching of whites/blues/reds/cyan/etc.

------

However, I will stand up and make a general statement:
A good LED system, using XP-G, and XP-E, with a decent spectrum, should look/grow the same at about 1/2 the power of MH.
And that does match up with what a LOT of others have done of RC and other places.

----

BTW:
If power used to heat your tank is a big issue:
1) Insulate. I laugh when I see these open-top, glass, no-insulation, tanks and sumps.
2) Use your gas hot water heater. Run a controlled recirculation loop through your sump.
 
rygh said:
Even in the not always useful lumens/watt, LEDs are 120-139, where MH are usually 70-85.
Data sheets numbers are pretty easy to measure. Not fudged.
Need to read them correctly though :D Cree XP-G run in the 110 Lumen/watt range, and that's only the cool white ones, and ONLY if you run them at 350mA, I know quite a few who over drive them to get more light, and the lumen/watt rating drops quite a bit. Halides can get upwards of 115 Lumens/watt, granted that doesn't apply to the bluer bulbs we use, but halides do have that kind of output potential.

Rating in mW is great, but how do you compare it??
So they ARE actually very efficient. Just hard to tell.
?? You don't know how to compare, as a result they must be efficient? Yikes man!

However, I will stand up and make a general statement:
A good LED system, using XP-G, and XP-E, with a decent spectrum, should look/grow the same at about 1/2 the power of MH.
And that does match up with what a LOT of others have done of RC and other places.
I'll make a general statement too, you probably can get away with much less wattage of halides and still get good growth rates than what people normally do as well :D

----
BTW:
If power used to heat your tank is a big issue:
1) Insulate. I laugh when I see these open-top, glass, no-insulation, tanks and sumps.
2) Use your gas hot water heater. Run a controlled recirculation loop through your sump.
[/quote]
Different strokes for different folks, up here in the blistery northern regions of San Francisco hot days are a rarity :D No need to use a hot water heater, I'll run halides and be happy with it :D
 
Look at data sheet again. XP-G, R5 = 139
Running at 700 mA, that drops to 126 or so. Ehh.

Look at how white LEDs are built, then think about why blue is likely very efficient.

115 lm/w for normal MH? You sure that was not some high pressure sodium or something.
I have never seen a standard MH bulb even remotely close to that. Even pure white industrial.
 
yes 139 lumens @ 350mA with a forward voltage in the 3.3-3.5V range (having gone through a dozen of these I can tell you I never got close to their "typical" 3.0V) that puts it at around 113-120 lumens per watt. Also remember these data sheets are strictly for the LED itself, not taking into account any losses by the drivers that are used, which can be significant considering the smaller number of LEDs per driver.

http://en.wikipedia.org/wiki/Luminous_efficacy
metal halide lamp 65–115 [36] 9.5–17%
however I do concede that these higher range bulbs would be far from anything we'd put over our tanks, as they'd be quite yellow.

Looking over the data for royal blues, and if I'm reading it right 350mW output says that at that voltage, you'll get that much light power out of it, so I'll concede they could be upwards of 30% efficient (raw) if I'm reading that right... again however the drivers can knock that down quite a bit depending upon which ones you're using, and how many bulbs you're using.

I'm still not sold on a 2:1 ratio of watt per watt comparison. Simply because I believe that while you CAN get away with it, you also CAN get away with a lower wattage of MH bulb as well, difference is there are distinct steps 150, 175, 250, 400 where as you can make the increments smaller with LEDs just due to their nature. I've been running LEDs over my anemone tank for a while and the anemones seem to be fine with much less power, however it could be a lot of spot lighting that allows them to take advantage of that (plus they're not terribly deep in the tank)
 
Just a question here, do mh and led produced light travel through water as well as each other?does one lose more intensity than the other as depth increases? Seems like they would be the same in that respect...
 
Mike: I'm not sure to be honest. I seem to recall someone saying lighting intensity for LEDs drops off a bit more dramatically because its a point source so all the light not going straight down will be at an angle, where as a MH bulb you'd get that same effect as well, however the MH will have the reflector that will spot light it downward (assuming good reflector) to reduce that. However I'm just hypothesizing and not sure in practice how it works.
 
Yes, the key to the depth issue is the angle of the light.
Light going at 45 degrees has to go 41% farther to get to the bottom than light going straight down.
Plus, with the rockwork, it is also likely to hit the sides of rock, and end up in shadow.
An LED with no optics has a roughly 120 degree spread +/- 60, so a lot of the light goes in at wide angles.
Although refraction of the water helps. It reduces the angle.
A MH light actually has a 360 degree spread, but it has a nice reflector around it, deflecting light fairly straight down.
The solution for LED is simple - optics. A lens will direct light down quite nicely.
Unfortunately, optics cost more, and you lose about 10% of the light in them.

Note that at wider angles above 60 deg, you also get reflection issues at the surface of the water, so
you do lose a bit of light that way.

The penetration of light does vary with wavelength, but that problem is the same for MH + LED.
 
I looked at that google page that says 115 lm/w.
It used Venture Lightings web site as a source.
But going through there detailed data, they use 86 as what appears to be the peak.
I never found anything real at 115, but might have missed it.
Looking a bit like marketing BS though.

Note that the google site also says 150 lm/w for LED, which is a bit high.

Yes, drivers have losses. As do MH ballasts.
My power supply is rated at 94% efficiency, and IF I build my own current sources, I can get 93% from a 48V boost-mode driver.
So 87% efficiency. I think some direct AC-> current are a bit higher, but not as usable for big systems.
I do not have data on MH ballasts.
 
I think there is no doubt that LED is more efficient than MH, esp. MH really suck in efficiency in blue segment. But I don't like many marketing BS of many LED companies I.E. claiming a 120W system is equivalent to 400W MH, that is simply misleading.

On positive side, I now tend to believe that LED is not lacking any spectrum after seeing many LED only SPS tanks, for example Oct RC TOTM is LED only tank, and he keeps tons of Acroporas with nice coloration.

Still there are something in Acroporas (only) under MH which is missing under LED or/and T5 lighting, the difference is hard to described & can only be captured by eye not by camera, I guess that is some sort of effect produced by corals when they fight with UV maybe?

Thus my upgrade I will shoot for 10K MH + heavy blue T5 & LED supplement.

Maybe in future I will give up & no more heavy on Acroporas, I will turn to LEDs.
 
ryanjiang said:
IOn positive side, I now tend to believe that LED is not lacking any spectrum after seeing many LED only SPS tanks, for example Oct RC TOTM is LED only tank, and he keeps tons of Acroporas with nice coloration.

Spectrographs don't lie, sorry. One just needs to look at the plot to see they do in fact miss a bunch of spectrums. There is no LED that hits all tha major spectrums, but MH's hit them all for the most aprt :D
 
Back
Top