Minimalistic multichip DIY LED build

O.K.., now I'm thinking that maybe all I need are 4 of the 50W chips that can be driven 1.7 - 3.5A. From what I'm trying to figure out, if I use a HLG-185-36, I could run two of these in Parallel at 5A putting 2.5 on each LED (right in the middle of their rated current). In the event one fails, what do I do to protect the other against over-current (3A fuse in-line on each string?)
 
Wow, 600 PAR that far down seems like a lot compared to 250 watt MH.

Yes it is a good result but the light is more concentrated because I use a 60 degree lens. To get an even light distribution, you have probably count on two 250 watt MH should be replaced with 3 styckien 100 watt LED

Sincerely Lasse
 
100 watts chip see #164. 10 cm above the surface, lens 60 degree. 10 cm below watersurface PAR 1500, 30 cm below watersurface PAR 1000 and 45 cm below the surface PAR 650.

With the 10 watts without lenses just under the modules PAR 900, 50 cm down - PAR 200. With lense much better

Sincerely Lasse

@ GY2202: Probably, I should take att least one or two 445 also.

Hi Lasse,
My water is 102cm/40" top of water to top of sand. If I am using 100W chips, would I need tighter optics than 60 degree to get say 300 par +/-. on the sand bed? and then would I also need to group the chips tighter together? Thanks---Rick
 
Hi Lasse,
My water is 102cm/40" top of water to top of sand. If I am using 100W chips, would I need tighter optics than 60 degree to get say 300 par +/-. on the sand bed? and then would I also need to group the chips tighter together? Thanks---Rick

I have not seen lenses narrower than 60 degrees for these chips. With my 100 Watt driven at 3000 mA, I got 200 PAR in air at 80 cm. Next week I will test with 3500 mA on the same chip - (you can get a current as high as 7000 mA on these chips). Will be back with the PAR values ​​of 3500 mA next week. Generally I tend to measure higher PAR in water than in air at the same distance. Probably due to the lens effect of the water.

I do not know how closely you need to put the chips to get good impact, I believe no one else knows either :). You just have to try


@ Dannym: With that set, I would invest in dimmable drivers - there is risk of it getting a little too well otherwise.

InDashMP3 said:
In the event one fails, what do I do to protect the other against over-current (3A fuse in-line on each string?)

Good idea.

Sincerely Lasse
 
Yes it is a good result but the light is more concentrated because I use a 60 degree lens. To get an even light distribution, you have probably count on two 250 watt MH should be replaced with 3 styckien 100 watt LED

Sincerely Lasse


Should be "pieces of"

Sincerely Lasse
 
Hi Lesse,

I did a few tests with my 10w LEDs on my meanwell 60-48p driver and they all work. My question for you is which one is better in your opinion. I've been looking at the LEDs direct during these tests. After that I can't see which option is brighter,safer, and more efficiency than others?

1) run 3 LEDs in series
2) run 4 LEDs in series
3) run 5 LEDs in series
4) run 6 LEDs in parallel (two strings of 3 LEDs)

As far as 4) option, what would be the resistor and fuse to use? From my calculation, I got 36w and 20 ohmn resistor and 80w fuse. Please let me know. I really appreciate it.

Thanks,
Ken
 
Last edited:
@ Haxer: The real difference is not the supplier but the fact that AC-RC's chip is 100 x 2.5 watts and is in reality a 250-watt chip if one uses 7000 mA current. At 3000 mA will give the chip 100 watts. The second chip is 100 x 1 watt LED and will not run at higher rates than 3200 mA and can only produce 100 watts maximum. If you only intend to use the 3000 mA so you equal with the difference that you run the AC-RC's chip at half its capacity (longer lifespan ?) And the second chip at maximum level (shorter lifespane ?)

@ kttran_2005: In cases 1-3, then current intensity across each chip be the setting on your drive. In case 4, the current across each chip will be half the setting on the driver. If you use the max (1300 mA), then each chip run of 650 mA. Even if you do not see any difference with your eyes there is a difference if you objectively measure the light intensity. I prefer to run my 10 watt chip with 900 mA because I think I get the optimum taking into account both intensity and lifetime. If you do decide to option 4, I think it is enough if you on each string secures with a 1 A fuse. You may, however, the less light output with case 4 if I have figured out it right :)

@ Landsailor: It depends on what distance you choose between the chip and the lens. With these lenses, you can vary the distance and thus the scattering angle. At least - they say so

Sincerely Lasse
 
@ Landsailor: It depends on what distance you choose between the chip and the lens. With these lenses, you can vary the distance and thus the scattering angle. At least - they say so

Sincerely Lasse[/QUOTE]

Lasse, I also did not know what the multi degrees in one optic meant. This is an interesting concept if you can buy one lens that will adjust to suit your needs. Only problem is from your post you sound skeptical to this?
 
I'm always skeptical of claim that is not consistent with what I learned in high school 45 years ago :) Of course you'll have more scattering of light if you move the lens closer or further away from the focal point which the lens is designed for. The disadvantage is that you do not get a clear focusing of the beam. You get no sharp distinction between light and dark and have a light leak that you might not want.

I have also discussed it with my son in law, he holds a PhD in optical design. Here, the high school student and the PhD entirely agree :)


But in a sense, it works probably with the demands you will have in an aquarium.

Sincerely Lasse
 
I understand what you are saying. I have not seen these, do you know if they are marked as to what degree you are using them at? I wondering if you could buy one to decide on which optic degree works best on a specific depth. It could save me buying 90 degree, when maybe I would prefer 60 degree? Or I guess I could buy one of each for testing.---Rick
 
The lenses I bought with a fixed number of degrees have all come with a reflector so the distance between the chip and the lens has been fixed, most likely, the gap was equal to the lens focal point. They give at least a sharply defined beam. I bought some "multi" lenses and they have a point that they are designed for. You can see that when you bring the lens back and forth. Now I only buy lenses with a fixed number of degrees - then I know what I get and I get a reflector with.

Sincerely Lasse
 
I thought I'd share because this thread got me interested in trying this for my frag tank I'm setting up. I picked up two 50w multichips a 10k+actinic and the other is a 20k+actinic. Tank is an 80gal (48l x 24w x 18h). It'll be interesting to see if I can get by with just the two 50w since the tank is pretty shallow or if I need to add supplemental lighting. Left hand side is the 10k and the right hand side is the 20k. I'm only using the reflectors and no optics at this time.
 
I used the driver recommended for the LED boards by that RC-DC guy:

50W Led driver for Actinic Blue Hybrid Led Panel
http://www.ebay.com/itm/22083316965...NX:IT&_trksid=p3984.m1497.l2649#ht_1721wt_905

85-265VAC 50W Dimmable LED driver 30-36V 1.6A
http://www.ebay.com/itm/27091599114...NX:IT&_trksid=p3984.m1497.l2649#ht_1208wt_905

I used these for the heatsinks:
http://www.newegg.com/Product/Product.aspx?Item=N82E16835200063&Tpk=Rosewill RCX-Z90-CP

Figured the heatsinks were cheap, copper base, and rated up to 95W so they should work fine.
 
Last edited:
Thanks for the info. What is your thoughts on the two chips, why not just use 2 20000k on dimming drivers?---Rick
 
Back
Top