DIY Multichip LED Build

[reza]

Thanks Lasse. What are the dangers in using this configuration?

[Lasse]

The danger is, as usual, that you must connected in parallel. In parallel connection current is divided equally between the channels. Do you feed six channels of 6700 mA then it is split into 6 * 1 117 mA / channel. Falls one channel it will be 6700 mA / 5 = 1340 mA / channel. The channels can handle 1400 mA so it will be no problem but if the two channels break, you also risk to burn the remaining channels. If you choose HLG-240H-42B (5.72 A) -you can lose 2 channels without much risk of burning the other 4.

If you're the cautious type, you should add a quick fuse on 1400 - 1500 mA in series with each channel.

The above is also in its way a response to 007Bond - surely it is possible to maximize the drives that you mention, but when it is on the border and it produce a lot of heat

@ 50 cube: The answer is - yes - it proves the math.

Sincerely Lasse

[ronreef]

Hi , so I'm planing on doing a build with 3 multi-chips on over a tank. I would like to connect all white together all blue together and violet together. I would prefer to use the least amount of drivers. Is it possible to get away with only using 3 drivers for the 3 colors?

Yes, I plan to run my lights that way as well. Only for the short term. You should have no problem given you have the right drivers.

Ron

[ronreef]

Forget the math, what's the answer?

The math provided the answer.

[007Bond]

The danger is, as usual, that you must connected in parallel. In parallel connection current is divided equally between the channels. Do you feed six channels of 6700 mA then it is split into 6 * 1 117 mA / channel. Falls one channel it will be 6700 mA / 5 = 1340 mA / channel. The channels can handle 1400 mA so it will be no problem but if the two channels break, you also risk to burn the remaining channels. If you choose HLG-240H-42B (5.72 A) -you can lose 2 channels without much risk of burning the other 4.

If you're the cautious type, you should add a quick fuse on 1400 - 1500 mA in series with each channel.

The above is also in its way a response to 007Bond - surely it is possible to maximize the drives that you mention, but when it is on the border and it produce a lot of heat

@ 50 cube: The answer is - yes - it proves the math.

Sincerely Lasse

Lasse, Here is a question for you guys in the know. I keep looking at solutions of pushing the Dream Chip towards its max because my tank is deep,(40" top of water to top of sand). Some are posting 15 to 20 inch deep tanks in which case I can see easily running their LED's at 700/800 mA.

I started thinking the ocean has less and less light as it gets deep,..if I'm try to put 200/300 par on my sand bed,.. will I end up cooking all my other corals on the way down to the 200/300 at the bottom? I mean will the coral in the top 3rd. of my tank fry at say a par of 700+ ? and I haven't even factored in optics? or am I not thinking correctly how light and water mixes? Thanks---Rick

[Bingles]

I got my 4 dreams.
I wont get drivers till thursday So I cant test them until then.

Thanks,
Brian

[ronreef]

I got my 4 dreams. I have questions about the dream build..
I have some concerns hope its no big deal but it doesn't look good under microscope. some channels are flawless builds others not so much.

1.) Few of my dream chips big solder gobs on the surface, the gold wires run over the top of them, doesn't appear to short anything or anything like that, not even sure if its solder but it coming out of the shield and onto the substrate where LED's are mounted.
*Clear photos would help us understand better what's going on.

2.) One one of my Dreams Two of the gold wires on negative side cross and secure to metal tab, others do no cross. I don't think this matters at all.
*Wouldn't worry about it.

3.) One of my dreams I have 2 single rows one complete channel has 4 gold wires that are glued down. There is two small globs but not near the gold wires and there is also the only channel that appears it took a few times to try and attach the gold wires, a lot tool scrapping on same pad. There is no tool scrapping on any dream chips that I have. The gold wires appear to be just sitting on the metal tab or maybe the gold itself is soldered but there is no gobs like all the other channels.
*Photos would help if possible.

The other 2 dreams appear to have good wire routing and and gold wires are attached plus glue. So my question is should I be concerned about any of what I mentioned?
*Cant really tell until you fire it up and test it out. When you say glue, do you mean some type of conductive epoxy?

Also did AC-RC doing any kind of basic test on them all?
*Yes, functional testing was done on all channels for all parts. I will reconfirm this with AC-RC.

I wont get drivers till thursday So I cant test them until then.
*Hang in there. Post photos if you can, and let us know how testing goes.

Thanks,
Brian

Please read my comments above. -Ron

[ronreef]

Brian,

Per AC-RC, all channels for all parts were tested and confirmed functional before shipping out.

Best Regards,

Ron

[Sonny]

Still waiting on mine to show up. I looked on my receipt for a shipper and didn't see one. AC-RC sent a picture of the package and there is a bar code with numbers, could it be a tracking number?

[ronreef]

Still waiting on mine to show up. I looked on my receipt for a shipper and didn't see one. AC-RC sent a picture of the package and there is a bar code with numbers, could it be a tracking number?

Probably best to email AC-RC directly.

[Lasse]

Still waiting on mine to show up. I looked on my receipt for a shipper and didn't see one. AC-RC sent a picture of the package and there is a bar code with numbers, could it be a tracking number?

Yes it is the tracking number. Go here

@007Bond I do not think you need to have the chip at max. You see the PAR values - it was without lens and only in air. The water surface will act as a lens. Also you have only around 8 cm from water surface to the LED. I´m not sure but if you do not get 200 PAR at the bottom - its no catastrophe - you can have corals there that not need so much light. If you run them at max you need a very good cooling solution.

Sincerely Lasse

[007Bond]

Yes Lasse, , I'm starting to think in that direction. I can see I will need "testing", and sounds like a par meter. But once again with the depth of my tank, if I'm getting some decent par at the bottom of the tank, will I end end up with to much light in the upper regions? In other words, if I push for 150 par at my bottom,( and with the depth ) could I in theory, have 1000+ par in the upper areas? I'm now worried more about too much par at the top, instead of to little at the bottom, all because of the depth. Need the happy medium. I seen Benighteds is suggesting to run at 4.5A.---Rick

[Lasse]

Yes Lasse, , I'm starting to think in that direction. I can see I will need "testing", and sounds like a par meter. But once again with the depth of my tank, if I'm getting some decent par at the bottom of the tank, will I end end up with to much light in the upper regions? In other words, if I push for 150 par at my bottom,( and with the depth ) could I in theory, have 1000+ par in the upper areas? I'm now worried more about too much par at the top, instead of to little at the bottom, all because of the depth. Need the happy medium. I seen Benighteds is suggesting to run at 4.5A.---Rick

Yes it will end up with high readings att the top but you can plan your "islands" that way that the highest point will not be in the center of the light beam. I think also it can be a good idea to try to get rails or something like that so you can move your light a little. My experiences is also that if you let the corals adjust to the ligt - they manage rather high PAR values.

Myself I wait just to see what Janne will come up with - with his solution you do not have to chose just now. they can be run up to 1400 mA. Have you decide how to run them. 20 channels, 10 channels or 5 channels? With HLG-185H-36B you can run 4 equal channels up to 1300 mA each but when you miss the possiblity to vary your light with invidual pendants.

Sincerely Lasse

[007Bond]

I'm thinking 5 channels combining 4 chips on 5- HLG-185-36B's and right now I have that DIM4 on my mind for controlling. It can control 4 channels, so I will have to double up 2 channels that will come on and power up and down at the same time, but I don't see an issue with that. Plus its cheap, and with its 2 on board relays I'll be able to turn on the cooler fans, and some type of over temp. shut down.(I think) I'm still want to see Janne's solution. Also I do have a rail and tilt solution in mind, but that will depend on the coolers. Speaking of which, are you testing your Zalman 8900? If it is compatible, and can be installed in the PVC pipe, it would make my rail solution much easier.---RICK

[Bingles]

I got 2 of my Dreams mounted on heat sinks with some nice fans, total cost so far was the price of the Dreams, arctic silver 5 ($7), and fans ($12), Drivers ($175 or so) heat sinks were free.... Still only about half or one third of what the cree diy kits cost and I wont have no "Disco" in my tank!!! And not to mention a heck of a lot easier to install, plus mine will be a on a rail system so I can slide and twist under the hood for best lighting.

One or two more days till drivers get here, I cant wait to see them over my tank!

Brian