Another Cree CXA3070 DIY Thread

zangtumtum

zangtumtum

Well-Known Member
in my project, for a space of 4x4 i use 9 COB at 1.4Amp,dimmable at 2.1,
in the calculation of the simulated light cones and distribution seems to be
the best solution,correct me if I'm wrong,
in air,only forced air :
Tamb 25C° = 40C° with 100 watt each, with 12" fan at 80cfm - 24db(i think real)
the heatsink can support more power obviously,verified with simulations and in the real application,
Peltier is mountable with other optional (entirely in copper,worked for liquid cooling...),
this is the crude heatsink,right now not anodized and not worked...
only "adjusted lapped" with CNC...perfectly flat,roughness 1,2 μm.
 

Attachments

Last edited:
P

PurpleBuz

Well-Known Member
epicfail said:
Those are peltier plates one side gets hot the other gets cold, you still need to put a heatsink on the hot side.

What @goud means by "put a next one on it" is use another plate to cool the hot side, its a joke can't you see the winky face?
Click to expand...
cool pic! ohh I get it now ... . Needing a heat sink and possibly a fan on the other side and the power requirements for the peltier seems useless.
 
SupraSPL

SupraSPL

Well-Known Member
The Peltier can be used to decrease the Tj but it requires electricity to do the job and it is very inefficient so it will actually create more heat in the grow space.

So the question is, does the decrease in Tj from the Peltier result in an increase in efficiency greater than the power used by the Peltier? In a 52W CXA3070 setup (22 PAR W), if we were able to reduce the Tj from 60C to 25C, the light output would increase by about 10% (2.2 PAR W) which would normally cost us 5.2W of dissipation. So if the Peltier uses 5.2W to do the job, it would be a wash and a loss overall because of slightly warmer grow space, more cost, more complex build and increased thermal resistance in the thermal path.

Of course it would take a lot more than a 5W peltier so unfortunately that tech is not ready for our application. Maybe if it becomes more efficient somehow someday?

There is a commercial lamp that attempted to do something along those lines but I cannot recall the name for sure, Kessil? They claim to use a vapor phase change heatsink which does not use electricity, but it represents an increase in complexity and increase in cost.

http://www.newegg.com/Product/Product.aspx?Item=N82E16835103189

This might be great for those who run their CPUs HOT, but because we run our COBs as close to ambient as possible, I suspect the vapor chamber system would not provide enough benefit to justify the cost. It would be more effective to spend the money running your COBs softer.
 
Last edited:
zangtumtum

zangtumtum

Well-Known Member
Peltier in our applications is inefficient in many aspect, but it is fair experiment, better to consider the right design in air or liquid, with heatpipe or mixed copper / aluminum with heat sinks for high power in order to remain below 50 ° C.
I tried to test Alpine 11 and 20 (better) but more than 80 watts must be cooled better and not by the fan 92 from the original that is not of quality and does not have the right pressure. Go over the 100watt for a long time would not do it with Alpine 11 or 20
I've noticed that being within or below 40/45 ° C degrees COB have a 10% increase in performance compared to those declared. Managing to keep them under 25C °, the increase is about 15% in lumen and par, measured by a professional appliance...imagine on liquid nitrogen...
 
P

PurpleBuz

Well-Known Member
zangtumtum said:
Peltier in our applications is inefficient in many aspect, but it is fair experiment, better to consider the right design in air or liquid, with heatpipe or mixed copper / aluminum with heat sinks for high power in order to remain below 50 ° C.
I tried to test Alpine 11 and 20 (better) but more than 80 watts must be cooled better and not by the fan 92 from the original that is not of quality and does not have the right pressure. Go over the 100watt for a long time would not do it with Alpine 11 or 20
I've noticed that being within or below 40/45 ° C degrees COB have a 10% increase in performance compared to those declared. Managing to keep them under 25C °, the increase is about 15% in lumen and par, measured by a professional appliance...imagine on liquid nitrogen...
Click to expand...
I understand exploration and experimentation but its a little silly to even think of applying a super liquid nitrogen cooling system for a simple cob light system. I shudder at the complexity and risk of failure requiring redundant systems.

nonetheless love to hear about your results for more data.
 
K

Kuifje76

Well-Known Member
Hey guys, i would buy a 20-pack cxa3070 AB's if I would be certain i could resell the other 12, i only need 8 for a 2'x4' closet.
Are there people here from Europe who would also like to buy and if so, what is the safest way these days to do such transactions ? Do i just take a picture of the receit to convince people it's the real deal (they're 'AB-bin")
 
The Erlking

The Erlking

Member
Kuifje76 said:
Hey guys, i would buy a 20-pack cxa3070 AB's if I would be certain i could resell the other 12, i only need 8 for a 2'x4' closet.
Are there people here from Europe who would also like to buy and if so, what is the safest way these days to do such transactions ? Do i just take a picture of the receit to convince people it's the real deal (they're 'AB-bin")
Click to expand...
Hi,
I could use 4 of them for my test build.
Pay-pal usually works fine and some pictures/documents that confirms the goods.
 
You must log in or register to reply here.
Top