Any LED DIY'ers build with these types ?
- Thread starter Jacob6364
- Start date
der_wille_zur_macht
Team RC
Having had a few LED fixtures over tanks for a few years, I can tell you anecdotally that corrosion is not an issue if the correct techniques and materials are used.
Another alternative is a high current driver running parallel strings. In the end, if you need "x" watts of LEDs, your driver choices are not really going to change because you switched from a bunch of individual LEDs to a modular component (that in itself is made up of many LEDs). You're still gonna be providing "x" watts to an array.
Again, it really depends on your criteria. These modular LED assemblies have been around more or less as long as HP LEDs in general, but they are not popular on fish tanks for a number of reasons. The main reason being the poor efficiency compared to what else is available. You're basically making a tradeoff - modular units offer poor performance but (depending on your point of view) "easier" design/assembly/maintenance.
Plus, you mention "saving money." IMHO it is important to qualify statements like that in threads about LED fixtures that might last 10+ years. Are you saving money right NOW, in terms of build cost, or are you saving money in operating costs over that 10 year period?
Given the rate at which LEDs are improving in terms of performance and cost, I am sure that within a few years there will be easy to use modular assemblies that are cheaper, more efficient, and offer better performance than today's "best" HP LEDs. At that point maybe it will be more common, but for now, people who have performance and efficiency as their primary goals are probably going to stick with individual LEDs. And since many of the people who are switching to LEDs are doing so because of the potential for improved efficiency over other forms of lighting, to them it doesn't make sense to choose an LED approach that does not optimize efficiency.
There are also the factors that are not easy to qualify with hard fact. For instance, flexibility - maybe you would like the color you get from one of those 100w modules, but other people might hate it. With a from-scratch build, if you don't like the color, you can swap out a few LEDs for a slightly different bin, or a different color alltogether. You have TOTAL freedom, whereas the modular approach locks you in to what the manufacturer offers off the shelf.
I hope my post does not come off as argumentative, I'm just trying to provide some context for your questions.
Since this is a cheap Epileds module, I would suggest you check into equivalent 3W LEDs. If you believe the specs you see online, you will find that its cheaper to buy individual 3W LEDs when you add up the luminous flux these things promise. Granted you need a somewhat bigger heatsink for individual emitters. The bottom line is you have to compare apples to apples. These cheap no-name LED emitters (the 50W or the 3W) are cheap for a reason. They don't guarantee output. People build tank lights with them -sure. I have one. But I would be very very cautious to put one of these over an established SPS tank for example. Its a gamble most people are not willing to take. When you see a beautiful SPS or even LPS tank that is not dead after 6 months using those cheap LEDs, then you know it works well. Otherwise you are just experimenting. Good luck!
Neogenocide
New member
I'll be honest here, I thought about trying one of these over my fuge, not really concerned about cost or efficiency, just curious to see how well they work. I've haven't done it yet, but I may in the future. It would be an interesting experiment to put a few frags under it as well.
The main reason I thought about trying one, less drilling, less tapping, less wires, less hassle for just lighting a fuge. And yes I've got 3 separate Cree setups currently running.
The main reason I thought about trying one, less drilling, less tapping, less wires, less hassle for just lighting a fuge. And yes I've got 3 separate Cree setups currently running.
der_wille_zur_mach,
no not argumentative at all, this is how information is gathered.
I do have a few questions for you though, throughout this evolution that we are seeing in the realm of LED's being used for reef aquarium, what is the age of the HP types?
are they a new thing ?
I am not at all doubting at this point that the cree XPE is the way to go, but i would love to see the HP modules evolve in quality, or even see the type that Firechild has that are built in blocks that could be a plug and play type.
please note that cost was never the top issue with my concern, it was modularity with quality not being discounted for that convenience, it would be nice to do the Pepsi challenge with these bulbs.
is there any reviews on these? i cant find any, just you tube stuff concerning the construction.
i would like to try a few setups for testing:
1. 100 watt hybrid with Blue 20,000K
2. 100 watt hybrid with blue 10,000K
3. 50 watt 10,000K-stand alone no blue
4. 50 watt 20,000k-stand alone no blue
take PAR readings on frag tank and measure over a 6 month period
originally my idea was to use the 100 watt 6500K for intensity at midday, so I made an excel cost study of 3 projects:
1. modular type
2. 1 soldering all the crees
3. 1 with the 100 watt 6500k
look and critique, I will work on adding to this which would include the hybrid types.
no not argumentative at all, this is how information is gathered.
I do have a few questions for you though, throughout this evolution that we are seeing in the realm of LED's being used for reef aquarium, what is the age of the HP types?
are they a new thing ?
I am not at all doubting at this point that the cree XPE is the way to go, but i would love to see the HP modules evolve in quality, or even see the type that Firechild has that are built in blocks that could be a plug and play type.
please note that cost was never the top issue with my concern, it was modularity with quality not being discounted for that convenience, it would be nice to do the Pepsi challenge with these bulbs.
is there any reviews on these? i cant find any, just you tube stuff concerning the construction.
i would like to try a few setups for testing:
1. 100 watt hybrid with Blue 20,000K
2. 100 watt hybrid with blue 10,000K
3. 50 watt 10,000K-stand alone no blue
4. 50 watt 20,000k-stand alone no blue
take PAR readings on frag tank and measure over a 6 month period
originally my idea was to use the 100 watt 6500K for intensity at midday, so I made an excel cost study of 3 projects:
1. modular type
2. 1 soldering all the crees
3. 1 with the 100 watt 6500k
look and critique, I will work on adding to this which would include the hybrid types.
havnmonkey
New member
From what I can gather so far, the OP really wants to give these LEDs a shot... unfortunately, so far, it doesn't seem as though anyone has tried these out.
I'd say go for it. It seems like you have the understanding and comprehension to take on such a feat and test it thoroughly. My experience is flashlights; and the one piece of advice I will share is this: when you (or manufacturers) combine LEDs into these small modules, overheating becomes one of the biggest limiting factor in their successes. Many flashaholics have placed multiple LEDs (XM-L, SSC-P7, Luminous SST-50 and/or SST-90) in the heads of maglites and can only turn them on at full power for a few seconds due to the inadequate heat dissipation.
I just built a 14 XP-G/XP-E build on a single AL U channel; at full power (limited at 750ma) the sink gets very warm but not hot. At 3/4 power it's barely above room temp. The heat sink is 36" long with each LED approx 2.5" apart.
I would make it a priority to heat sink these 20/30/50 Watt modules well.
I'd say go for it. It seems like you have the understanding and comprehension to take on such a feat and test it thoroughly. My experience is flashlights; and the one piece of advice I will share is this: when you (or manufacturers) combine LEDs into these small modules, overheating becomes one of the biggest limiting factor in their successes. Many flashaholics have placed multiple LEDs (XM-L, SSC-P7, Luminous SST-50 and/or SST-90) in the heads of maglites and can only turn them on at full power for a few seconds due to the inadequate heat dissipation.
I just built a 14 XP-G/XP-E build on a single AL U channel; at full power (limited at 750ma) the sink gets very warm but not hot. At 3/4 power it's barely above room temp. The heat sink is 36" long with each LED approx 2.5" apart.
I would make it a priority to heat sink these 20/30/50 Watt modules well.
Neogenocide
New member
I would agree with what havnmonkey said about cooling, maybe consider a CPU heatsink, something with a copper core and heatpipes to help wick away the heat. That is what I plan on using if I ever get to that project. In fact I already have one sitting here. My Radio Controlled passion is starting to take over again, so my funds are being diverted that way or else I would lead the charge. www.newegg.com is a great place to find the heatsinks that I'm talking about.
der_wille_zur_macht
Team RC
I think there's some difference in terminology here. When I say "HP" I mean high power, as in pretty much any LED designed to operate above 1W. All the LEDs in this thread qualify.
FWIW there are a handful of manufacturers that are basically doing what firechild has done - making MCPCBs that hold many "good" LEDs such that you only have one connection point for several emitters. Check out cutter in Australia (who refer to them as "light engines"), and I think even some of the reef-specific vendors are doing this now.
Thing is, there's always a tradeoff - if you want better modularity you're gonna lose in either cost and/or performance. Even if these manufacturers were starting with the best emitters in the world (and arguably epileds is not) then they'd lose performance by stuffing a few dozen on one die. Neogenecide commented on this, and it's easy to see elsewhere. Look at the Cree MC-E, which is literally four of the same emitter used in the Cree XR-E, but built on one die. The performance figures are basically prorated because of this, you can't run it at the same currents and you get less lumens per watt.
Again though this is not terminal - the modular LEDs will probably work fine, but at a slight performance hit, and that's probably OK given your priorities.
That would be a very interesting study indeed.
tahiriqbal
Sialkot
Sorry about the late reply Jacob6364, been busy with work.
When I started to play with these LEDs (three years ago) I was mostly interested in the Kelvin rating rather than PAR readings as company was claiming to have them @20K plus. IMHO, when I tested them in the labs they came out no more than 8000K at which point, I thought they may have made a mistake!!!! I informed the company about the results only to be told that they sent me the wrong LEDs (Big Lie) and will resend new LEDs. The next set of LEDs arrived which company tested before sending them out as 20K plus and guess what :thumbdown lab results showed them @ 8000K or there about (AGAIN). I sent the results back only to be told that they don't really make 20K LEDs but they are cool enough to be called 20K LEDs. I think this was very misleading on company's behalf and would not trust these HP LED manufacturers hyped up claims.
Also, failure rate on these LEDs was very high and no suitable lenses were available and most importantly very inefficient when compared with lumen per watt. Here are few pictures which may give you some idea on the experiment;
Another project of mine....:love1:
Tahir
der_wille_zur_macht
Team RC
Depends on how you want to structure the experiment. I suppose your "control" could be an array built to hVe the same theoretical output. Though as I alluded to in my first post, it isn't really the sort of thing you need to determine experimentally because you can look up the relevant specs on paper.
An interesting point. When each emitter is in it's own chip, a single emitter failure is cheap and easy to fix. When you have 20 or 40 emitters on one chip, a single emitter failure means you throw away 20 or 40 emitters.
I am going to buy four of their p30s to replace my 2x 400w halides.
How are their LEDs???
Similar threads
