DIY watercooled photosynthetic-research LED-Lamp completed build, tell me what u think
- Thread starter l0wbob2016
- Start date
l0wbob2016
Well-Known Member
well you dont need to add 12 different monos haha, also i dont think u want any wavelength to be dimmable. so it would be one extra driver and some more wire and its done. this goes quicker than u think
l0wbob2016
Well-Known Member
good luck fixing that COB on that heatsink without damaging or glueing it. ( by glueing i mean thermal paste with 2 components that harden out )
Danielson999
Well-Known Member
l0wbob2016
Well-Known Member
well after this build i could clearly say that using the AiO-watercooling solution i found works very good for a CXB3590 driven with 1400ma constant current for about 4 hours in a non ventilated room.
once i have done the testrun in my friends room i can tell more about the performance, but for now i'm pretty sure the radiators wont get warmer than 40°C which means the COB's may be at about 45°C?!
i think thats a pretty good value if u take a look in the datasheed of these COB's when it comes to electrical characteristics in combination with chip-temp.
in this case :
Datasheet:
http://www.cree.com/led-components/media/documents/ds-CXB3590-1.pdf
Page 11 to 13.
in the very worst case i would hit that red line, but i should mostly performing better. if needed i will take some values and check it with math whats going on
if i got anything wrong please tell me.
once i have done the testrun in my friends room i can tell more about the performance, but for now i'm pretty sure the radiators wont get warmer than 40°C which means the COB's may be at about 45°C?!
i think thats a pretty good value if u take a look in the datasheed of these COB's when it comes to electrical characteristics in combination with chip-temp.
in this case :
Datasheet:
http://www.cree.com/led-components/media/documents/ds-CXB3590-1.pdf
Page 11 to 13.
in the very worst case i would hit that red line, but i should mostly performing better. if needed i will take some values and check it with math whats going on
if i got anything wrong please tell me.
Very cool setup. Doubt Id recommend that people copy the design, but it is pretty awesome.
Just dont think it really takes advantage of any of the perks watercooling setups offer.
But I do like the ability to customize the spectrum with the lighting.
Out of curiosity, how much of the $1600 went towards the cooling? A quick google search listed those CPU coolers at $100 a pop. If you spent ~$500 on cooling I think anyone here that has toiled with LED design can say thats not a cost effective solution.
Just dont think it really takes advantage of any of the perks watercooling setups offer.
But I do like the ability to customize the spectrum with the lighting.
This is a fact. Its not that it wont work, its just that a passive setup could probably be done more cheaply, which is kinda funny when you think about it.
Out of curiosity, how much of the $1600 went towards the cooling? A quick google search listed those CPU coolers at $100 a pop. If you spent ~$500 on cooling I think anyone here that has toiled with LED design can say thats not a cost effective solution.
Last edited:
l0wbob2016
Well-Known Member
i never wanted to be cost effective. i wanted to know how good these coolers perform on the complete build using them as a cooling option for the COB's & the LED's. also i can "dimm" the waterpumps and the fans down if needed ( with the driver ) to about 25-30 W. there isnt much less cooling when dimmed down so i let that setting.
since i got in total 5 COB's and 60 mono's that get cooled down to ~ 45°C i save some power. sure it may not be as much as 25-30W but i got in theory all COB's and LED's "actively" cooled for under 30W. so it may not be much efficient but this is the configuration i wanted.
What do you mean that you "save some power" by cooling them to ~45c?
I guess my point is, there is no advantage to watercooling in this setup, so why use it when there are better and cheaper ways to do it?
My DIY watercooling build can cool ~360w of LED using about ~20w of cooling power and only cost about $100.
Its a moot point with regards to this light, but you did mention it was a prototype. If you build another on the same cooling design it would simply be foolish.
nogod_
Well-Known Member
This argument has never made sense and still doesnt. The way youre interpreting the inverse square law is positing that somehow photons are vanishing into thin air.
In a closed room everything reaches or is reflected back down onto the canopy. If the effect you are trying to achieve is even color mixing, the higher you hang your monos, the better.
In a closed room everything reaches or is reflected back down onto the canopy. If the effect you are trying to achieve is even color mixing, the higher you hang your monos, the better.
Yodaweed
Well-Known Member
That's wrong, that's why mono panels without lenses should be hung within 12 inches. Photons do not escape but they spread out when you hang your lights high, that's why PAR ratings diminish the further away from the light they are taken. Mono's do not produce enough intensity to be hung at the same height as COBs, unless you put lenses on them, then they just make tiny hot spots.
l0wbob2016
Well-Known Member
the cooler a COB or LED is while at 100% power ( for me here it is 1400ma ) the more light is produced. if i dont got anything wrong from the datasheets than driving for example a CXB3590 ( 36V@1400ma) with a temp of 25°C gives about 70% relative luminous flux. at 55°C about 65%, at 85°C about 60% and at 105°C about ~55%?!
lets say just for example i am at 55°C 36V 1400ma 65% relative luminous flux. to get the same relative luminous flux at 85°C 36V it needs about 100 - 150ma more ( ~1500-1550ma ).
Power is P = U x I;
[55°C 50,4W per COB] compared to [85°C for 100ma more: 54W for 150ma more: 55,8W] so thats 3,6 to 5,4W per COB i save in power when i want to have the 65% relative luminous flux from the chip.
if anything is wrong pls tell me! i always want to learn!
maybe for u there is no advantage, maybe for others to but for me it is. again this never should be cheap or cost effective.
lets just say i save 3W per COB with that calculation above, times 5 are 15W less powerdraw for 65% relative luminous flux from the COB. lets assume i save 0,05 to 0,1W per mono-LED, times 60 are 3 to 6W. so in worst case i get 18W less powerdraw for the same relative luminous flux from the complete build.
l0wbob2016
Well-Known Member
i got 90 degree lenses on the monos. and 60 monos in total. u are right the further u go away from the source the less useable light comes down on an area especially when the source isnt that powerfull like a COB, but these monos were never ment to be the only light source the plant gets, they just should support on certain wavelengths.
i know that they wont throw much light compared to the COB's
Yodaweed
Well-Known Member
Ok that makes sense , thanks for the clarification.
nogod_
Well-Known Member
Dude takes his first crack at crafting an LED lamp and makes a great looking one out of quality components, with a tuneable spectrum, for someone else to use, and all you can think about is the cooling system.
Folks need to take a deep breath and pick your battles.
Bottom line: I'd love to play with this lamp and you'd have to be an idiot to hangout in this section and not feel the same.
Folks need to take a deep breath and pick your battles.
Bottom line: I'd love to play with this lamp and you'd have to be an idiot to hangout in this section and not feel the same.
VegasWinner
Well-Known Member
My experience the 3w mono's work just fine at the same elevation as the COB.s. Light travels the same speed in short distances under a mile at 86,000 fps. So a couple of feet light arrives at or near the same time. The intensity may be lower for the mono's but sill effective. Plants are looking for red and blue not intense red and blue. I think you will find out you may dim your cobs, but probably not your mono's. Do you have InfraRed in your spectrum? If not you might consider adding it, does great things to plants.
What exactly is the advantage then? People make passive setups that are as effective for cheaper... Its not like watercooling is actually benefiting you in this case over an aluminum heatsink and fan because thats essentially the same thing, but far cheaper.
He did ask for comments and questions... I commented it appears he overspent by atleast a factor of 2 to cool his light. I asked him why.
Yeah, agreed. But Id also say its good to learn from mistakes, not double down defending them lol.
Im still very curious to see someone with a cob/mono setup run a grow with monos and without to gauge the difference.
Partly what I said about what I liked about this particular light, but when you criticize any part, people just assume youre a hater, regardless of validity or not.