Designing new led

I've been building my own led fixtures for some years now even a couple liquid cooled units that preformed very well actually, anyways I work in a very nice cnc machine shop with capability to build anything and I come up with some new ideas I think could greatly improve efficiency and overall quality. I've noticed over the years led efficiency is dependent on several factors, temperature, spread, even your driver. I don't necessarily think liquid cooled is gonna be the next big thing but it cannot be overlooked, I'm wanting to try surface mounting diode directly to the heatsink eliminating the mcpcb all together. Like aluminum plate with copper fins that go all the way thru where the thermal pad of the led can be soldered to and flat copper wire for the circuitry. I was also thinking of some benefits of liquid cooling like super thin fixtures for racks or adding thermoelectric coolers for a/c there is so many possibilities. I would like to know what yall would like to see from future lighting fixtures. I'll post pictures of some of my other fixtures later today

Jonesfamily7715 said:

I've been building my own led fixtures for some years now even a couple liquid cooled units that preformed very well actually, anyways I work in a very nice cnc machine shop with capability to build anything and I come up with some new ideas I think could greatly improve efficiency and overall quality. I've noticed over the years led efficiency is dependent on several factors, temperature, spread, even your driver. I don't necessarily think liquid cooled is gonna be the next big thing but it cannot be overlooked, I'm wanting to try surface mounting diode directly to the heatsink eliminating the mcpcb all together. Like aluminum plate with copper fins that go all the way thru where the thermal pad of the led can be soldered to and flat copper wire for the circuitry. I was also thinking of some benefits of liquid cooling like super thin fixtures for racks or adding thermoelectric coolers for a/c there is so many possibilities. I would like to know what yall would like to see from future lighting fixtures. I'll post pictures of some of my other fixtures later today

Click to expand...

Here is the one I'm working on right now
2 cmu2287 high cri cobs 1 3000k and 1 4000k
24 xpg3 660nm
16 xpg3 royal blue

Added some more diodes to the mix 16 xpg3 photophyll 2.5 gbr to the fixture and I am gonna consolidate this other light I was working on to beef this thing up a little 24 more xpgd photophyll diodes and 16 more xpg3 photo reds on 10 heatsink usa stars the stars will be ran passive cooling. I'm gonna be adding a couple drivers and figure out a final wattage I'm shooting for. I have ir on a seperate fixture and 54w pure uv t5 bulbs I'm wanting all these colors on separate channel to do my own investigation into what spectrums bring the results I am looking for vs what I've read / been told the cobs have a 240w rating but best efficiency is 82w or lower iseen the data sheets on these cobs and w9nder why I've never seen anyone try flowering with these, considerably cheaper than cxb3590.
















.

Just use the money and effort spent to get more chips, running at lower current.
You get better spread, more efficiency, less expense and more free time.

What kind of photon losses would one expect if they just submerged the LEDs in thin glass containers of mineral oil, setup like an RDWC system with bulkheads that hangs from the ceiling? Lets say they are right near the glass (or acrylic/etc), with only a thin layer of the oil allowed to flow in between, which is cooled with a remote mounted radiator, and pumped through the whole system the whole time it's on.. I been wanting to know this for years now, but never forked out for a light meter to do some simple tests like that.. Glass and oil above your plants, what could go wrong?

Drop That Sound said:

What kind of photon losses would one expect if they just submerged the LEDs in thin glass containers of mineral oil, setup like an RDWC system with bulkheads that hangs from the ceiling? Lets say they are right near the glass (or acrylic/etc), with only a thin layer of the oil allowed to flow in between, which is cooled with a remote mounted radiator, and pumped through the whole system the whole time it's on.. I been wanting to know this for years now, but never forked out for a light meter to do some simple tests like that.. Glass and oil above your plants, what could go wrong?

Click to expand...

What's the point? The front side of the package is pretty well isolated with the silicone lense and the back side is directly soldered to the aluminium pcb.

I started doing this when led where just getting big there wasn't very many good fixtures back then. Epistar trash. I do think lm301h is over rated and there are better options but I'm not looking to make a profit. I understand that for the price it's the best option, for a company looking to make a profit it's hard to pay 2-3 $ per diode. And have you seen the diffrent flux bins available? Just because a light says samsung chips doesn't mean they are top bin diodes. 5hat can change the efficiency dramatically.

The oil is non conductive of course, so you can immerse them directly in to transfer the heat away. Instead of pumping a water based solution through channels\pipes. It's also clear depending on the type used, so should let a majority of the light shine through, especially with the chip resting near the bottom towards the plants. Why are data centers and massive bitcoin operations turning to immersion cooling if its not worthwhile? Even future home computer concepts appear to have multi phase immersion coolers, with crazy new technologies built in.. Same with future LED grow lights i'm sure. I think the future of grow lighting will be laser beams that are projection mapped to the plant as it grows, sensing every leaf movement. Scanning so fast you cant tell its a few single beams, and only using the exact amount of light that is necessary.

I guess the point is to drive chips past their limits for fun, or make them last a long time, and mostly to create or find new ways to grow a plant? See if there is anything the last guys that tried it missed out on. Its already been done the way I described with mineral oil, a few times I know of..Not with any reliable follow up tests utilizing light meters or anything I've seen though.


It's fun tinkering around and coming up with new innovative designs!







^ Would like to see better screw in bulb tech.

Never thought of fully emerging the led in liquid but I could see that working I have some 2' rails I built a while back they work great just a bitch to set up I have a copper coil made from 50' 3/8" copper tubing hang that thing up in the corner and just loop it thru the light and pump. I was just always scared of pump failure. But there is flow switches for pretty cheap that will fix that

I wanted to try something like the above design I thought of at the time, but newer LED strip lighting tech was coming out (hundreds of tiny COBs per meter), and I figured I could just adhere the long strips directly to a single copper pipe once they released better full grow spectrum versions...

Have you seen the spectrum of these 95+ cri cobs I chose them because looking at the data sheet the higher the cri the farther out in the red spectrum they go. I was thinking of laying them on some 1 1/4 " square aluminum pipe and capping the ends with barbed pipe fitting and running water thru em. Maybe even drilling a 2" hole outside my shop maybe 15' deep so I can bury copper pipe for a vertical ground loop to geothermal cool my fixtures and I want to play around with some hvac to use the cooling that way as well

Here's another one for you ... we use propane burners for co2 correct? Why don't we just burn anthracite? It burns very clean as well its actually approved for indoor fuel. Have you seen how long that shit burns? Wanna build a tiny coal based heater for winter if I get around to it.

I've been wanting to put radiant floor heating coils in my rooms, and use an on demand portable propane water heater in a closed loop to heat it with. I was thinking, I could heat the floors in the cooler months, and also use it like a co2 burner at the same time! Thinking about it, the on demand water heaters are more like a water cooled co2 burner, and you could use it that way too, to get c02 with a burner but without the extra heat.

I have about 80 lb tub of anthracite coal outside somewhere, that came from the old mines around here before my time.. Huge chunks! Not even sure what I was saving it for, probably incase I wanted to setup a small foundry or something..

Slapping another few fixtures together I've been using these copper 4 up XP dtp mcpcb from MTN electronics for years they do run these diodes a lot cooler at higher currents. It is important to note electricity is cheap where I live I think it's like .08 ¢ per kWh so efficiency is not my main concern more like playing with different colors to see what works best. This one's got 72 xpg3 whites 48 3200k 80 CRI and 24 5700k 80 CRI, 16 xpg3 photo red 4 xpe2 far red and 4 xte blue. So 96 high power diodes frankenlight I'ma run around 280 watts

I cant remember the name of the company, but i think the guy who runs it his name is vitali, growmouse used to work with him. They been putting out optional liquid cooled grow lighs an optional air cooled grow lights for some time. Im sure they made one of the most efficent grow light e ever too. Some good designs. Oh yeah i think their called chilled tec led.
Liquid cooled can be very effective, will be cool to see some of what you come up with.