Smoke-A-Cola
Well-Known Member
Can someone please tell me what kind of grow tent this is?
Water cooled COBs
- Thread starter loftygoals
- Start date
ttystikk
Well-Known Member
Nothing special- and kinda weak, considering the roof is bowing that badly under the weight of a few LED lights.Can someone please tell me what kind of grow tent this is?
The light fixture is bowing, not the tent.
I think quictents is what youre looking for @Smoke-A-Cola
Last edited:
loftygoals
Well-Known Member
Yeah the light fixture was bowing as it was originally made from 2mm thick, 1m long painted steel flats. I have since reinforced them with Aluminium angles and there is now no bowing:
The name of the tent can be found in a few of my previous posts, I believe they are UK only...
(http://green-qube.com/roof-qube/)
ttystikk
Well-Known Member
Ahhh, gotcha. I see it now, wasn't so easy to spot before. My water cooled modules are made from 2"x4" by 3/16" thick wall extruded aluminum bar stock- stuff that's suitable for structural material for bleachers, pedestrian bridges, etc.
Before you say, 'overkill', I've quite literally dropped them onto a concrete floor (by accident!) and they were undamaged. They are heavy, though; easily 4-5 kilos each full of coolant.
mahiluana
Well-Known Member
Nugachino
Well-Known Member
Not to thread jack. But, I'm interested in running 1-3 cobs on a Waterblock.
I've got an old Thermaltake water kit for pc. Never been installed. It's been tested once. But that's all.
My question is. What would it take to cool between 120 and 270w. That's running a couple CXB's at least %70 power.
I'm not sure how to word this. So, sorry if I'm not asking clearly. But, I was thinking of swapping the single 120mm radiator for at least one 240mm rad. Maybe 2. Depending on what you guys think.
Initially I was thinking of running one cob @120w by itself. But then I thought I should probably make it at least two to share the power load more efficiently.
Cheers in advance
I've got an old Thermaltake water kit for pc. Never been installed. It's been tested once. But that's all.
My question is. What would it take to cool between 120 and 270w. That's running a couple CXB's at least %70 power.
I'm not sure how to word this. So, sorry if I'm not asking clearly. But, I was thinking of swapping the single 120mm radiator for at least one 240mm rad. Maybe 2. Depending on what you guys think.
Initially I was thinking of running one cob @120w by itself. But then I thought I should probably make it at least two to share the power load more efficiently.
Cheers in advance
mahiluana
Well-Known Member
Hi brother - since I bought a heatexchanger I will never pay for radiators and fans again.
You just need a little waterpump(2,3W), hose + fitting and a closed watercircuit.
You can find cheap combinations of heatexchanger and waterboilers on ebay:
http://www.ebay.de/itm/381284978091?_trksid=p2057872.m2749.l2649&var=650518629979&ssPageName=STRK:MEBIDX:IT
little water pump:
http://www.ebay.de/itm/322157900813?_trksid=p2057872.m2749.l2649&ssPageName=STRK:MEBIDX:IT
All together I payed ~200,- EURO.
In the moment I run only 300 W - led-lamp, but it produce enough hot water for 2 persons
(30 gal/ day) and I`m able to save ~50,- EURO in my heating bill.
The combination of heatexchanger and waterboiler has some advantage.
- flexibility in case when you need more than one shower.
- there is normaly a magnesium-anode giving protection against corrosion.
- you can heat up the system once or twice a year to kill bacterias.
A solution without heatexchanger can be a big waterbucket and an old socket
specially if you want more humidity in your growroom
The amount of wasted heat-energy by led-light is still 70-80% and do not depend so much upon the way how you drive your chips. Heat production of 1 chip @ 1500mA is more or less the same like 3 chips @ 500mA.
But light-efficiency (lumen/w) is much better if you drive your chips @ low current.
Buying a triple chip-set and running it @ 33% current can increase light-efficiency 40-50%.
If you know type and flux of your chip you can compare here:
http://pct.cree.com/dt/index.html
It`s a question of money if you invert in light-efficiency - don`t forget that longlife
of your chip will rise with lower current.
Anyway - you should be able to get back ~ 150 W from your 270 W lamp with watercooling.
This is also the dimension of heat you want to combat with radiator and fan
Using hot water from a watercooled led lamp can rise lamp-efficiency > 100%
- because 25% light and 75% heat means that a watercooled led-lamp is much better a (water)heater than a (lamp)lighter.
DIY can bring this together.
ttystikk
Well-Known Member
Your assertion that LED lights still emit 70-80% of their output as heat is incorrect. They're mostly over 50% efficient at conveying Watts into PAR, and some combinations exceed 70% efficiency.
While it's true that you can reuse the heat elsewhere in your home (I'm doing it myself, in fact), it's not the main output of the chips. This stands in direct contrast to HID lighting, where most of the output is indeed heat.
mahiluana
Well-Known Member
Hi Ttystikk
- ok there is a lot of difference depending the chips quality`and also driving them.
Records in efficiency are mostly made in manufactures laboratories and I guess it`s hard to measure and discus exact values because eg. light or photons are converted to heat when they impact on a surface. Led chips for consumers like you and me don`t reach 70% efficiency.
You are very optimistic
250%Somewhere I read about a 683 lumen / watt maximum of a green 555nm theoretical led chip,
which is a physical frontier of efficiency converting 100% of the input power into light.
With very, very good 200lumen/watt you would be still around 30% light efficiency
I measured the lower part of a cheap chin. 50W cob chip running at full power ( 33V@1,5A ).
2 chips @ 750mA each produce ~10% more light and a bit less heat in my cooling water.
I was able to store 160W from a real 295W lamp with 6 cobs in my cooling water.
This is 54% of inverted electr. energy and even more, if you add the energy loss on the way to the thermometer - and please don`t forget the upper side of the chip which is hard to measure.
Conclusion in my case would be - less than 45% of inverted energy is light and heat emitted from the upper side - in theory I split it up in 25% light and 20% heat to complete my bill.
Thinking and drinking I imagine a little cpu watercooler connected 100% by welded copper to the deepenings of the upper surface of the led chip - like a dentist`s tecnician can do while copying your teeth.
Amount and temp. of water then give you an idea what kind of energy flows up there.
In an optimized (watercooled) transport of the heat away from the led chip I measured a logical and significant lower phosphor temp. on the upper side (~30°).
Soon I will do the same measuring with cree chips (CXA2530/5000K/T4) and meanwell drivers - I guess there will still be ~45% of stored heat energy in my cooling water.
Later I tell you more.
SORRY - in a oneman company this can take time - see you
Last edited:
ttystikk
Well-Known Member
@REALSTYLES built one and he's not a manufacturer's lab flunky! Read up on current droop.Hi Ttystikk
- ok there is a lot of difference depending the chips quality`and also driving them.
Records in efficiency are mostly made in manufactures laboratories and I guess it`s hard to measure and discus exact values because eg. light or photons are converted to heat when they impact on a surface. Led chips for consumers like you and me don`t reach 70% efficiency.
You are very optimistic
Somewhere I read about a 683 lumen / watt maximum of a green 555nm theoretical led chip,
which is a physical frontier of efficiency converting 100% of the input power into light.
With very, very good 200lumen/watt you would be still around 30% light efficiency
I measured the lower part of a cheap chin. 50W cob chip running at full power ( 33V@1,5A ).
2 chips @ 750mA each produce ~10% more light and a bit less heat in my cooling water.
I was able to store 160W from a real 295W lamp with 6 cobs in my cooling water.
This is 54% of inverted electr. energy and even more, if you add the energy loss on the way to the thermometer - and please don`t forget the upper side of the chip which is hard to measure.
Conclusion in my case would be - less than 45% of inverted energy is light and heat emitted from the upper side - in theory I split it up in 25% light and 20% heat to complete my bill.
Thinking and drinking I imagine a little cpu watercooler connected 100% by welded copper to the deepenings of the upper surface of the led chip - like a dentist`s tecnician can do while copying your teeth.
Amount and temp. of water then give you an idea what kind of energy flows up there.
In an optimized (watercooled) transport of the heat away from the led chip I measured a logical and significant lower phosphor temp. on the upper side (~30°).
Soon I will do the same measuring with cree chips (CXA2530/5000K/T4) and meanwell drivers - I guess there will still be ~45% of stored heat energy in my cooling water.
Later I tell you more.
SORRY - in a oneman company this can take time - see you
wietefras
Well-Known Member
That's not how that works. Lumen doesn't measure all light equally. It depends on the spectrum.Hi Ttystikk
Somewhere I read about a 683 lumen / watt maximum of a green 555nm theoretical led chip,
which is a physical frontier of efficiency converting 100% of the input power into light.
With very, very good 200lumen/watt you would be still around 30% light efficiency
Lumen is measured with a peak at the 555Nm wavelength while other wavelengths further away from that are counted less in the total lumen count. That does not mean the light is not emitted. Just that it's counted less in the lumen measurement.
For an exact 555Nm light 683lumen/W would indeed be 100%, but for a 3000K Cree 328lm/W would be 100%. Worst case, if all the light falls outside the 380Nm to 780Nm range of the Lumen curve, the led could be 100% efficient and still measure at 0 lumen/W.
So lumen is only a measure for efficiency when you have the LER for that particular light source (683lm/W or 328lm/W or whatever).
mahiluana
Well-Known Member
Do you think driving Cree`s 328lm/W there is no heat production at all ?I also had been surprised by 70% light efficiency as mentioned by @ttystikk and i still think
it`s around 15-35% in comercial products or diy led lamps.
That`s why i said that a watercooled led light is much more a water heater than a light source.
The step from a generator to produce electricity ---> to a generator that produce
electricity + hot water was the same. From 30% in 1970 up to 75% today.
In general lighting sucks ~ 25% of an average, global electricity bill.
If 30% of lighting would be watercooled by heatexchanger the planet should be able to burn 4% less fossiles and that is a huge number
I don`t want to watercool your bed-side lamp, but i guess there are plenty of places where
you can find simultaneous demand for bigger quantities of light and heat.
ttystikk
Well-Known Member
You still have your numbers wrong in terms of output efficiency, but the concept you're talking about is well known; cogeneration is the idea of using multiple byproducts of a given industrial process such as electricity production.
I also had been surprised by 70% light efficiency as mentioned by @ttystikk and i still think
it`s around 15-35% in comercial products or diy led lamps.
That`s why i said that a watercooled led light is much more a water heater than a light source.
The step from a generator to produce electricity ---> to a generator that produce
electricity + hot water was the same. From 30% in 1970 up to 75% today.
In general lighting sucks ~ 25% of an average, global electricity bill.
If 30% of lighting would be watercooled by heatexchanger the planet should be able to burn 4% less fossiles and that is a huge number
I don`t want to watercool your bed-side lamp, but i guess there are plenty of places where
you can find simultaneous demand for bigger quantities of light and heat.
For example the light from COB LED is being used to grow my plants. The heat is used to heat my home and domestic hot water. This is cogeneration at work.
I'm excited about the promise of natural gas powered fuel cells since every byproduct of its use can be utilized; water, CO2, heat and of course electricity.
mahiluana
Well-Known Member
That`s what i mean - i simply separate light- from heat-production.
Watercooling enables a pretty exat measuring of the transfered heat.
I will measure top efficient leds and drivers for their heat output soon - then i`ll tell you.
keep your grow up vertical - cheers
wietefras
Well-Known Member
Yes, that's the definition of 100% efficiency. They won't ever reach that, but still. 70% is expensive since you need to run the COBs very soft (and/or have an imaginary super cooled system and imagine that increases efficiency by 20%), but 50% is easily attained with DIY.
How about trying to understand the subject before making bold (yet untrue) statements?
Shit, this tech has been right around the corner for about two decades now... dont know if Id get too excited if I were you. Havent kept up with the tech lately though, so maybe it will become more than vaporware in the not-so-distant future.
mahiluana
Well-Known Member
@wietefras Do you agree, that most of comercial led products still have processed low efficient (20-30%) leds ?
wietefras
Well-Known Member
What's the point of that claim? What does it matter if I do or don't agree to it?
The point was that you claimed it's impossible to build or buy anything better than 30% efficient, because you made the incorrect assumption that you need 683lm/W for 100%.
We mostly do DIY here and those are usually somewhere around 40% to 50% efficient. Of course you can buy crap commercial led fixtures, but there are plenty available in the same 40%-50% range.
MeGaKiLlErMaN
Well-Known Member
I'm not sure about my future lights efficiency... But driving a 130W cob at 17.5-20W will produce a lot more efficientcy. Having around 48 of them in a 4x4 area will provide the best lighting coverage I've seen. I would hope to see over 60% efficiency at least. We will see the LED calculator shows around 55% but it also doesnt have these cobs. Anyone have that QER calculator? It had the graph included I think.. Might not be the correct name...