wow.. some of that made sense.... well most of it lol... but I'm still lost as to how big the heat sink needs to be.. I think 3 and 3 leds would work... do you think I should do 6 blue and 2 white pc or like 3 white led 3 blue led 2 white pc.. or some other line up lol.. and it's for the bio
DIY LEDs - The write-up
- Thread starter JohnL
- Start date
- Status
Soundwave,
I'm planning out a build very similar to yours for a 75 gal and I was curious how hot that heat sink gets, and if there may be a less heavy duty option that would help keep the weight down.
What about just using two 9.5" x 3" heat sinks on each side? Each of the four heat sinks would have 18 LEDs on it.
Also, based on what you have said previously, I was thinking about using 72 total LEDs and going for some sps. I figure it it is too much I could just raise or lower the fixture.
Thoughts?
I'm planning out a build very similar to yours for a 75 gal and I was curious how hot that heat sink gets, and if there may be a less heavy duty option that would help keep the weight down.
What about just using two 9.5" x 3" heat sinks on each side? Each of the four heat sinks would have 18 LEDs on it.
Also, based on what you have said previously, I was thinking about using 72 total LEDs and going for some sps. I figure it it is too much I could just raise or lower the fixture.
Thoughts?
Last edited:
Reef_Matrix29
Water Lover
The heatsink needs to be big enough to house all of your LED's and give about 1" of space around all edges for good cooling and support of the array, IMO. Well, if you are doing 6 LED's, my question would be are you trying to replace your existing lights or compliment them? Basically, what you are suggesting so far is complimenting them and really you will still be replacing PC's every 6 months or however long you do it....so you aren't really saving too much money. However, if you are looking to replace your lights and depending on what you are keeping in the tank, I would do either 8 or 12 LED's in a 1:1 ration of blue to white and wire them to dimming drivers so you can adjust color pushed @ 700mA with no optics if you will have them relatively close to the water line.
:wave:
Reef_Matrix29
Water Lover
I have a 75 gallon and I will tell you that 72 LED's 6 inches off the water line is NOT too much, but I'm pushing mines @ 700mA, so you could increase that if you choose. Not sure what the benefits would be to have a 9.5" x 3" heatsink would be since w/ the additional weight comes better heat spread and protection for your LED's which are the most expensive part of the DIY. Keep in mind that the array itself won't be the huge weight, its the canopy or support system that could be culprit. I experienced the same thing, but my total canopy/array weight is 24 lbs roughly!
diy led setup
diy led setup
I'm looking at doing a led setup,but first can you give me a ball park on cost and savings on electric vs the setup I have now.Tank is 24"x30"x96" is all sps and the lighting 4 - large lumenbright fixtures,2- dual 250w coralve ballast,12k reeflux bulbs.I'm looking for a 14k light spectrum.Thanks for your replies
diy led setup
I'm looking at doing a led setup,but first can you give me a ball park on cost and savings on electric vs the setup I have now.Tank is 24"x30"x96" is all sps and the lighting 4 - large lumenbright fixtures,2- dual 250w coralve ballast,12k reeflux bulbs.I'm looking for a 14k light spectrum.Thanks for your replies
der_wille_zur_macht
Team RC
24x30x96 is 299 gallons. Surface area 2,880 square inches, depth 24 inches.
You currently have 250w MH x 4. Those ballasts are probably pulling at least 300 - 325 watts. We'll call it 300. So, 1200w of power. Assuming 8 hour photoperiod, 9.6kWH of electricity per day.
You could probably get similar light from ~250 LEDs. Let's assume Cree XR-E Q5 bin, for the sake of argument. 700mA, wide optics. That's around 2.45w per LED. If we assume drivers and power supplies 80% efficient (being conservative), that's 3W each. So, 750 watts. Again, assuming 8h photoperiod, 6.08kWH per day.
So, the LEDs will be around 37% cheaper to run, in terms of kilowatt-hours. Depending on your electricity cost, that'll translate differently to $$$. I think I'm around $.12 per kWH, but I'd bet the country varies from .07 to .40 depending on where you are. Let's say $.15. Each day, your cost will be about $.42 less, or $12/month, or $144/year.
Doesn't sound like much, considering the LED rig probably costs twice what the MH rig does, but think in the long term. Over 10 years, you're saving $1440 on electricity. If the cost of electricity rises 5% a year, it's a lot more than that. Plus, bulb changes. If you change those four bulbs every 8 months at $80 a pop, that's 60 bulbs, at a cost of $4,800!
So, your 10 year savings would be around $6,240, or $624 per year. If the MH rig costs you $2500 to make (pulling this from you-know-where, I'm out of math for this post), you'd pay it off in 4 years. If you had been starting from scratch and comparing the initial cost of the MH rig, the payoff would probably be more like 3 years.
And, it's arguable that the LED rig could be constructed more efficiently than I'm showing - fewer LEDs, if you position them carefully (according to your coral placement, for instance). Fewer LEDs again, if you use the new XP-G LEDs. Combining these factors, you could probably be 30% - 40% more efficient again, which would double the electric savings in the long term.
Of course, you could also argue that playing with an MH rig, it could be made more efficient - but I think it's pretty clear that over a long period (10 years) the LEDs win, and they'd pay for themselves in ~2 - 5 years.
You currently have 250w MH x 4. Those ballasts are probably pulling at least 300 - 325 watts. We'll call it 300. So, 1200w of power. Assuming 8 hour photoperiod, 9.6kWH of electricity per day.
You could probably get similar light from ~250 LEDs. Let's assume Cree XR-E Q5 bin, for the sake of argument. 700mA, wide optics. That's around 2.45w per LED. If we assume drivers and power supplies 80% efficient (being conservative), that's 3W each. So, 750 watts. Again, assuming 8h photoperiod, 6.08kWH per day.
So, the LEDs will be around 37% cheaper to run, in terms of kilowatt-hours. Depending on your electricity cost, that'll translate differently to $$$. I think I'm around $.12 per kWH, but I'd bet the country varies from .07 to .40 depending on where you are. Let's say $.15. Each day, your cost will be about $.42 less, or $12/month, or $144/year.
Doesn't sound like much, considering the LED rig probably costs twice what the MH rig does, but think in the long term. Over 10 years, you're saving $1440 on electricity. If the cost of electricity rises 5% a year, it's a lot more than that. Plus, bulb changes. If you change those four bulbs every 8 months at $80 a pop, that's 60 bulbs, at a cost of $4,800!
So, your 10 year savings would be around $6,240, or $624 per year. If the MH rig costs you $2500 to make (pulling this from you-know-where, I'm out of math for this post), you'd pay it off in 4 years. If you had been starting from scratch and comparing the initial cost of the MH rig, the payoff would probably be more like 3 years.
And, it's arguable that the LED rig could be constructed more efficiently than I'm showing - fewer LEDs, if you position them carefully (according to your coral placement, for instance). Fewer LEDs again, if you use the new XP-G LEDs. Combining these factors, you could probably be 30% - 40% more efficient again, which would double the electric savings in the long term.
Of course, you could also argue that playing with an MH rig, it could be made more efficient - but I think it's pretty clear that over a long period (10 years) the LEDs win, and they'd pay for themselves in ~2 - 5 years.
Thankyou for the reply.My main concern was heating issues that I had around july thru aug. running it without a chiller.Running central air,has helped. It has been a cooler summer, than normal in Michigan.How many leds in white and royal blue ,would I be looking at for this setup.Looking for a 14k light spectrum.
der_wille_zur_macht
Team RC
Indeed. If the 1200w MH system and the 750w LED system are putting the same light out (being rough here) then using the MH system is like running a several-hundred-watt space heater in your hood.
I would start at 50% blue and 50% white. Get dimmable drivers, and tweak in small amounts to get the color you want. Just don't tweak more than a few percent, or you'll be cutting light intensity pretty severely.
I would start at 50% blue and 50% white. Get dimmable drivers, and tweak in small amounts to get the color you want. Just don't tweak more than a few percent, or you'll be cutting light intensity pretty severely.
I'm not worried about the efficiency. I just want a boost the lighting.. should I do all blue leds and add a white pc or do all white leds and keep the blue pc.. what would make more light
Reef_Matrix29
Water Lover
Well, I guess if you want a "boost" in light only and aren't worried about any other benefits of LEDs....just add some white LEDs and keep the blue PC's, not sure how you will position them both within your hood, but I'm sure that you will work something out.
der_wille_zur_macht
Team RC
You need a controller that can put out a signal that can be translated to what the drivers want.
The dimmable buckpucks want a 5 - 0 V DC signal (5 is full off, 0 is full on). I don't know of any off-the-shelf reef controller that can do that, though some of the high-end controllers can do something that could be translated to that range - Apex, Profilux, etc - these high-end controllers can do a 0 - 10v DC signal typically. So you could use software on the controller to map it to 0 - 5 V DC over time to dim, simulating sunrise or sunset.
A reefkeeper lite is NOT going to do this. Note that the controllers that can do this are all $500 - $1000.
If you are handy with electronics, you can DIY a controller for this purpose for about 10% of that, at most. I've got one working based on the Arduino, but it's not quite ready for public release. PM me if you want details.
der_wille_zur_macht
Team RC
Honestly, if dimming and efficiency aren't important, just adding another PC would probably be easiest and cheapest (cheapest from an upfront perspective. LEDs will nearly always be cheaper than anything comparable in the long term.)
der_wille_zur_macht
Team RC
It'll take some effort, but lots of people have posted PAR numbers in this thread. I think at one point someone even posted numbers a few months after install, and saw that there was something like a 2% drop vs. brand new - though this is probably well within the range of accuracy for instruments used by hobbyists.
Reef_Matrix29
Water Lover
+1
Agreed!
Agreed!
Reef_Matrix29
Water Lover
I would be very interested to see what the PAR readings in my tank are, but I personally don't know anyone in the Baltimore, MD area that has a unit to take measurements and its not in the budget just yet with a wedding being planned now.
:hmm3:
Reef_Matrix29
Water Lover
Uh, yeah! With plenty of light to spare @ 700mA
- Status
