20K does have less par but one of the reasons people sacrifice par is the need not to have actinics along with the associated cost. Most people I know that run 10k also run actinics so that variable should come into play when comparing mh/led. So, side by side I would compare 20k 400mh with the latest led straight up. If I were to compare 10k mh with led I would include the cost of actinics with the mh. Seems to me that having the heat directed away from the tank should keep the tank temp lower, as opposed to the direct heat of mh. For the record I'm not sold on led's yet but the AI version does have me interested.
Solaris LED Lights
- Thread starter Lucifa68
- Start date
20K does have less par but one of the reasons people sacrifice par is the need not to have actinics along with the associated cost. Most people I know that run 10k also run actinics so that variable should come into play when comparing mh/led. So, side by side I would compare 20k 400mh with the latest led straight up. If I were to compare 10k mh with led I would include the cost of actinics with the mh. Seems to me that having the heat directed away from the tank should keep the tank temp lower, as opposed to the direct heat of mh. For the record I'm not sold on led's yet but the AI version does have me interested.
P.Kelly
New member
What no one seems to be posting is that LEDs do not create Infrared radiation (according to Sanjay Yoshi). Infrared radiation is the major contributor to heating the water in our tanks under metal halides. So, to say they give off the same amount of heat, but not directed into our tanks is a little misleading. Clear benefit to LEDS on this one, now for the price and reliability...
Hi guys,
I will not argue with you but I have my system running for 2 months now and I can give you some true details:
- My system is the PFO I-4 48" on my 125G tank.
- I use the "Auto Solar" mode with the color temperature set on 70% of the daylight leds and 50% of the blue leds. I also worked with 100% of the daylight in the beginning and reduced it later.
- I don't use chiller.
- The temperature of the body never excided 95 and most of the time is around 90. The fans work in a manner that most of the time only one pair is working.
- The temperature of the water is very stable, 77 - 80.
- The room temperature is not changing more than 0.5 degree during day time.
All that are facts.
I live in Israel, near Jerusalem, and I think that the climate condition of Israel is close to that of California.
Moshe
I will not argue with you but I have my system running for 2 months now and I can give you some true details:
- My system is the PFO I-4 48" on my 125G tank.
- I use the "Auto Solar" mode with the color temperature set on 70% of the daylight leds and 50% of the blue leds. I also worked with 100% of the daylight in the beginning and reduced it later.
- I don't use chiller.
- The temperature of the body never excided 95 and most of the time is around 90. The fans work in a manner that most of the time only one pair is working.
- The temperature of the water is very stable, 77 - 80.
- The room temperature is not changing more than 0.5 degree during day time.
All that are facts.
I live in Israel, near Jerusalem, and I think that the climate condition of Israel is close to that of California.
Moshe
buck50bmg
In Memoriam
Thank you! Looks like a 12year old did the wiring. Plus its cheap china labor and materials. I think all PFO does is import them and take the little gold "china" sticker off and sell them.
It was a slap in the face when I found out they were not US made. Then to find out not all the functions work on the controller and they are not going to do anything about it. Well for the money spent thats not good enough.
Then the LED strips that keep burning out. I did some research and looked at it from an electrical engineer prospective and found the way they set up the power supplies was just cheap and irresponsible if they want any longevity of the LEDs.
The Luxeon LED need a steady current source to operate. Using computer power supplies and expecting a stable voltage source/ current source is crazy. The LED have power modules that are to be used with them. Phillips the maker of the luxeon strongly recommend them as well. Weird that other users of the LEDs are getting a full life of the LED, but PFO is somehow burning them up.
InsaneClownFish
New member
Gomer where is your support for that statement?
buck50bmg
In Memoriam
Gomer hate to do this.......but....
How much of the LED light emitted is a) radiation b) efficient light?
How efficient is an LED in full operation?
No harmful ultraviolet or infrared radiation is contained within the Luxeon LED light beam. Because Luxeon LEDs are directional (not emitting light in all directions as in typical conventional light sources), high fixture efficiencies of 80% to 90% can be achieved, therefore requiring less total lumens to provide the same level of illuminance.
The efficiency of the LEDs in terms of lumens per watt output, depends on the colour. Colours towards the red end of the spectrum will have a larger light output than those at the blue end. Typically, a 1 watt white Luxeon LED has a light output of 25 lumens per watt.
http://www.lumidrives.com/Faqs.aspx?id=4
Amphibious
New member
Thank you, buck50bmg, for clearing that up with facts. That link was very valuable, too.
Yes
All the light is emitted as radiation by definition
It is pretty efficient in terms of PAR. We aren't quit there in terms of max lumen extraction, but we have made big jumps recently. We are just under the 50% threshold for white conversion. RBG's can hit a much higher lumen/watt mark.
The IR I was refering to is in the electron to photon conversion waste energy which easily gets pulled off by efficient heatsinking.
And this is WAY WAY outdated. Cree R2's are pushing 100lumens per watt.
Have you seen a solaris opened?
Have you built numerous high powered LED light systems?
I have.
I don't have a link to the inside of a solaris. It was posted in a "repair"thread by someone sometime back.
pjf
Premium Member
Solaris claims 80 lumens per watt from its white LEDs (http://www.solarisled.com/).
Aquaillumination claims 100 lumens per watt from its white LEDs (http://www.reefcentral.com/forums/showthread.php?s=&postid=11096385&highlight=bin#post11096385).
Is there any information to verify or rebut these claims?
Aquaillumination claims 100 lumens per watt from its white LEDs (http://www.reefcentral.com/forums/showthread.php?s=&postid=11096385&highlight=bin#post11096385).
Is there any information to verify or rebut these claims?
Their claims are probably true, but they don't tell you at what current 
I am pretty sure the new solaris usesthe new K2s. Aquaillumations uses Souls
As you increase current, you decrease efficiency at the expense of brighter light.
The "bright" bins of K2's are 95 lumens at 350mA, 170 at 700ma, and 275 at 1500mA wth a Vf of 3.3, 3.5, 3.85 respectively. This translates to 82 lumens/watt, 70 lumens/watt, and 48 lumens/watt. So the solaris is 80 lumens/watt under conditions that the solaris DOES NOT run under
I bet y ou'll get the same results with aquaillumination on the Soul LEDs.
I am pretty sure the new solaris usesthe new K2s. Aquaillumations uses Souls
As you increase current, you decrease efficiency at the expense of brighter light.
The "bright" bins of K2's are 95 lumens at 350mA, 170 at 700ma, and 275 at 1500mA wth a Vf of 3.3, 3.5, 3.85 respectively. This translates to 82 lumens/watt, 70 lumens/watt, and 48 lumens/watt. So the solaris is 80 lumens/watt under conditions that the solaris DOES NOT run under
I bet y ou'll get the same results with aquaillumination on the Soul LEDs.
buck50bmg
In Memoriam
Gomer
I am really glad you are helping demystify the LED issues. At some point in the future they may be economical, but right now they are NOT.
I like you hate to see people get ripped off. If we dont do the research ourselves or have a place like this to get information we need then the companies that sell this stuff are going to win, for some people a $4000.00 light for softies makes sense....but for others they need the information you helped gather.
As I have said before, the quality of the fixtures and the price just don't jive for me. I know the cost of the LED are high, but cold solder joints and shooty engineering are unacceptable in my opinion.
Im positive that PFO overdrives the LEDs. I remember looking at the fixture and thinking....all they did is put the LEDs in series and use the LED to regulate current....... They have drivers for a reason.
I think PFO should be ashamed of the solaris fixture.
I am really glad you are helping demystify the LED issues. At some point in the future they may be economical, but right now they are NOT.
I like you hate to see people get ripped off. If we dont do the research ourselves or have a place like this to get information we need then the companies that sell this stuff are going to win, for some people a $4000.00 light for softies makes sense....but for others they need the information you helped gather.
As I have said before, the quality of the fixtures and the price just don't jive for me. I know the cost of the LED are high, but cold solder joints and shooty engineering are unacceptable in my opinion.
Im positive that PFO overdrives the LEDs. I remember looking at the fixture and thinking....all they did is put the LEDs in series and use the LED to regulate current....... They have drivers for a reason.
I think PFO should be ashamed of the solaris fixture.
cclough_AI
Premium Member
Gomer:
You've brought up a great point. Companies like Seoul and Cree (we use both) state that they have 100 lumen/watt emitters. This is idealized though, and this efficiency is found at ~1W (350mA). This can be different though depending on what bin of part you receive (Vf and Min Luminous Flux)
For example, while the Seoul P4-U is spec'd at 240 lumens at 1A, this actually amounts to ~70 lumens/watt. This is of course dependent on the Vf of the emitter.
We use the Q4 bin Cree part and Seoul U bin (both considered 100 Min Luminous Flux) for the white emitters. Cree has a Q5 bin out now, which has slightly greater output, but they have huge lead times, and the small increase doesn't justify their significant price increase.
As for driving of the LEDs, in our case we use constant current buck regulators. We drive the white LEDs at ~1000mA and the blue LEDs at ~750mA, which for both manufacturers is within spec.
Hope this helps.
-Chris
You've brought up a great point. Companies like Seoul and Cree (we use both) state that they have 100 lumen/watt emitters. This is idealized though, and this efficiency is found at ~1W (350mA). This can be different though depending on what bin of part you receive (Vf and Min Luminous Flux)
For example, while the Seoul P4-U is spec'd at 240 lumens at 1A, this actually amounts to ~70 lumens/watt. This is of course dependent on the Vf of the emitter.
We use the Q4 bin Cree part and Seoul U bin (both considered 100 Min Luminous Flux) for the white emitters. Cree has a Q5 bin out now, which has slightly greater output, but they have huge lead times, and the small increase doesn't justify their significant price increase.
As for driving of the LEDs, in our case we use constant current buck regulators. We drive the white LEDs at ~1000mA and the blue LEDs at ~750mA, which for both manufacturers is within spec.
Hope this helps.
-Chris
Cree actually is on the R2 now
I can get any of them right now. There are no lead times on the crees if you know where to get them (Cutter)
I do think LED technology is good enough right now. It is just a bit pricey. Cree just released their 4 died LED which is basically 4 crees stuffed into an LED slightly larger then their normal ones. Soul has a 4die but it is large and wired parallel. Cree's had indvidual leads per die.
Every time I see the solaris, I am tempted to "do it right", but I just don't want to invest the money lol.
I can get any of them right now. There are no lead times on the crees if you know where to get them (Cutter)
I do think LED technology is good enough right now. It is just a bit pricey. Cree just released their 4 died LED which is basically 4 crees stuffed into an LED slightly larger then their normal ones. Soul has a 4die but it is large and wired parallel. Cree's had indvidual leads per die.
Every time I see the solaris, I am tempted to "do it right", but I just don't want to invest the money lol.
cclough_AI
Premium Member
Gomer -
We have some of the R2, but aren't currently able to get them reliably stateside, so I'm not considering that to be an option (yet). Cutter is in Australia, correct?
As for the multi emitter, single die parts that Cree and Seoul have - they're neat (I really like the Cree part) but the efficiency isn't greater than the single emitter/single die parts. Granted, we could switch from the current XR-E parts to the MC-E and run at 350mA each rather than 1000mA to effectively gain efficiency and output, but we're still consuming more power from the wall and the cost would go up significantly.
-Chris
We have some of the R2, but aren't currently able to get them reliably stateside, so I'm not considering that to be an option (yet). Cutter is in Australia, correct?
As for the multi emitter, single die parts that Cree and Seoul have - they're neat (I really like the Cree part) but the efficiency isn't greater than the single emitter/single die parts. Granted, we could switch from the current XR-E parts to the MC-E and run at 350mA each rather than 1000mA to effectively gain efficiency and output, but we're still consuming more power from the wall and the cost would go up significantly.
-Chris
You are correct, the efficiency hasn't changed at all, however it does raise prospects for a more "point source" tye of array. I have a feeling that there is going to be the next sizable jump in efficiency within the next year.
As for cutter, yes they are in australia, but a ton the serious Cree hobbiests in the US get them from there. Service is good and as with turnaround time.
I am tempted to do a setup using the 6 cree boards using a Q5 cool white x3, blue, and 2 of the older 400nm UVs. The mix should be a nice ~14k ish look. No it won't have color tunability, but I don't care about that. Given that cree's have a native 90deg full angle spread (vs ~120 on others), you can probably get away with no optics if the LEDs are near the center of the tank and not elevated way up. You can run 2 boards per common 700ma driver. 3 boards + 3 drivers = ~ 45 watts and will likely output an intensity similar to a 70watt MH (given my previous experience with 6white Q5s). Alternately if you ca
As for cutter, yes they are in australia, but a ton the serious Cree hobbiests in the US get them from there. Service is good and as with turnaround time.
I am tempted to do a setup using the 6 cree boards using a Q5 cool white x3, blue, and 2 of the older 400nm UVs. The mix should be a nice ~14k ish look. No it won't have color tunability, but I don't care about that. Given that cree's have a native 90deg full angle spread (vs ~120 on others), you can probably get away with no optics if the LEDs are near the center of the tank and not elevated way up. You can run 2 boards per common 700ma driver. 3 boards + 3 drivers = ~ 45 watts and will likely output an intensity similar to a 70watt MH (given my previous experience with 6white Q5s). Alternately if you ca
