Water cooled COBs

Discussion in 'LED and other Lighting' started by loftygoals, May 19, 2016.


    ttystikk Well-Known Member

    Heating oil is generally much cheaper per BTU than electricity, but it is an energy savings for sure.

    Shugglet Well-Known Member

    You going the insulating route or simply running your system at a higher temp? With a dialed in setup like yourself I wouldnt think it would be that difficult to figure out the dewpoint and run your light cooling system a degree or so warmer than the dewpoint.

    MeGaKiLlErMaN Well-Known Member

    It's easy to avoid, if you keep your room temp at say 75F then keep the water around 70F or so and you're golden. Just a different way to look at it. Less worry about it heating up your room and more lukewarm water.

    ttystikk Well-Known Member

    Turns out that a warm room that aids transpiration is desirable.
    MeGaKiLlErMaN likes this.

    1212ham Well-Known Member

    The blocks will be machined for a rubber seal. It's not something the average person will be doing at home.
    Would someone like to offer some insight to the questions I asked?

    Shugglet Well-Known Member

    The specs sheets from the manufacturer should list mounting specifications for thermocouples.

    But to be honest I see very little gained from having the water in direct contact with the COB vs a standard cooling block. Reason being that water is the weak link in the thermal conductance chain. Maybe Im wrong but it seems to be maximum work for minimal gain.

    Shugglet Well-Known Member

    You just reworded the question I wrote with arbitrary temperatures and ignored the insulation route... Unsurprisingly ttystikk responded to you with a quip of common sense that addresses nothing commented on. lol

    mahiluana Well-Known Member

    This way you have to add extra heat energy to your grow room - to stay in opt. conditions,
    because the colder water sucks air temp. down. - In summer under the roof maybe a good idea,
    but condensation will be an issue - tty says

    The other way round it`s easier.

    I measured in my grow room eg. : air 78°F / coolingwater 82°F at state of equillibrium.
    300W lamp / 40L water bucket in a windy place. You have to add water to the bucket
    every few days, if you run it that way.


    Shugglet Well-Known Member

    You dont have to add heat, you just remove less. Condensation only becomes a problem if your cooling system is below the dewpoint of the grow room. His scenario will almost certainly never result in condensation forming. How do you cool your grow?

    mahiluana Well-Known Member

    This is a 80L heatexchanger. Together with a coolmac-300W watercooled led lamp cost you

    700,-US$ incl. ship. + duty. :bigjoint:

    15h/day :idea: generates me ~90L warm (86°F) water.


    Shugglet Well-Known Member

    So just an extraction fan to actually cool the room? I mean, I understand how your lighting cooling works. Im referring to the grow room.

    mahiluana Well-Known Member

    2 x 120mm fans + carbon filter (6W / DIY) push out very few air. - I don`t have a heat problem.:weed:
    ttystikk likes this.

    mahiluana Well-Known Member

    Yesterday i did a very strange, "exciting" test to know this a bit more accurate.

    I wired a Cree 2530/ 5000K/ T4 to a 10W @ 300mA led driver.
    I wanted to be shure to use a middle class quality chip and run it under low current efficiency.

    My wattmeter showed 10,9W / Vf = 32,4V @ 300mA.

    All calculations of temp. change and heat energy you can do here:

    I wrapped the chip into a waterfilled "ballon" 0,175L and switched the light on.
    After 30 min. the temp. rised from 25,4°C to 40,5°C ---> 6,2W during 1800 seconds.

    I estimate the power loss for the led driver ~ 0,5W

    So 10,9W(powermeter) - 0,5W(led-driver) - 6,2W(heat energy) leaves 4,2W for the light.
    Light efficiency is 38,5%.

    Today i repeated the same test, but only 20 min. - I wanted to know more about the heat loss
    or heat transfer to the ambient.

    After 20 min. the temp. rised from 28,4°C to 38°C --->5,9W during1200 seconds.
    Following the water temp. of the ballon cooling down in the room i measured 9 min.
    to take the temp. down from 33,7°C to 32,7°C. ---> 1,2W / 9min. = 2,6W / 20min.
    This must be more or less my average heat transfer loss to the ambient and is bigger than i thought. You have to add it to the bill.
    :cuss:UUUUOUPS - that`s bad for light efficiency.

    New calculation is: 10,9W total - 0,5W led driver - 8,5W heat energy --->
    my light efficiency is just 26,6%.

    Any coments and critic is welcome - on the pics you see simple things for the measurement,
    that you can find in smart lighted household.
    I would like to see many results from different chips - with different spectrum and power.
    Can anybody help me please ! :bigjoint::dunce::fire::dunce::bigjoint::idea::idea::idea:;)

    Raging Stalk and ttystikk like this.

    ttystikk Well-Known Member

    Whoa- science, dude!
    mahiluana likes this.

    Schalalala Active Member


    Alter, jetzt mal ernsthaft.... haben sie dir auf der Förderschule nicht gesagt das du nicht mit Strom spielen sollst?
    In deinem Sicherungskasten haste aber nicht rumgeforscht oder?
    Falls doch, dann ist die Wahrscheinlichkeit das du dich mittels dem oben gezeigten "Prüfstand" umbringst gar nicht mal so gering... also hau rein, mach doch noch mal ne Messung, z.B. mit deinem pinken 1800K COB :mrgreen:.

    mahiluana Well-Known Member

    Duennschiß yako wewe kiduchu katika kinywa - pengine kuchoka na schwachmaaten nyingine ya grower.ch (suaheli für anfaenger)

    Schalalala Active Member

    Junge, mach es nicht noch schlimmer und lass den peinlichen Scheiß.
    Ich hab tatsächlich eine Frage zum "Versuchsaufbau":
    Liegt die LED auf Bild 2 flach auf dem LED-Treiber und das Kondom dann auf der LED? Oder ist die LED im Kondom?

    VegasWinner Well-Known Member

    Check out this water cooled cob with cheap ebay cobs.
    mahiluana likes this.

    Shugglet Well-Known Member

    @mahiluana Something you might be interested in.

    Also, estimating your driver efficiency could throw your final numbers pretty far off... Are you sure your driver is ~95% efficient?

    Also, your revelation about your heat loss to ambient is why I made my previous comment about using smaller cooling blocks.

    Also, in regards to you losing heat to the ambient air, your system is going to have far different characteristics of heat escape than your test setup. You really need to conduct your tests for your losses to ambient on the system itself. Case in point, your light bars will likely radiate heat far better than a water filled condom.
    Last edited: Mar 16, 2017

    mahiluana Well-Known Member

    Hi brother :peace: my main interest was to proof the enormous heat production of led light.

    Because many people still believe that led light is the new tecnic with low heat production.
    That`s not true. :wall: @ttystikk @wietefras

    Finally this morning i found a statement of a prominent member:

    http://www.cree.com/~/media/Files/C... Application Notes/XLampThermalManagement.pdf

    page 5 - heat generation.

    So for the future i hope there will be less confusion. (:

    Back to the numbers - that means every led light will be a light source with < 25% light efficiency.

    *Coolmac watercooled led system* is so far the only way to rise efficiency a huge step forward.

    25%:idea: ---------------------------------------------------------~~~> 85%:fire: ----------:wall:

    next target will be a watercooled led lens - hunting the remaining 15-20% heat power, wich is still emitted by LES and upside the chip.

    bicit likes this.

Share This Page