DiY LEDs - How to Power Them

U

UniversZero

Active Member
So, I may have grossly overestimated the passive cooling abilities of my bars. After running for an hour or so, these guys get pretty hot... Not so much that I can't keep my finger on them, but seem to be much hotter than I would have thought... Certainly above 85° by feel alone.IMG_20160320_175103.jpg So, here's what I'm running...
4 cxb3590 3500k (72v) on a meanwell hlg-185-700a. The sink is the 7.28 profile from heatsinksusa at a length of 24". I'm guessing that im losing all kinds of efficiency here, and possibly might damage the CoBs at full power?

Clearly, I'm going to need some fans, so would one large central fan be enough, of should I put two equidistantly placed on each bar? Any recommendations?
 
alesh

alesh

Well-Known Member
NM High Desert said:
Driver output was 169v on my multimeter with the pot disconnected, any input as to what I should be seeing?
Click to expand...
Well that's exactly the value I'd expect with this setup. Could be that watt meter is off...you can measure the current in the LED circuit to confirm.
 
Greengenes707

Greengenes707

Well-Known Member
UniversZero said:
So, I may have grossly overestimated the passive cooling abilities of my bars. After running for an hour or so, these guys get pretty hot... Not so much that I can't keep my finger on them, but seem to be much hotter than I would have thought... Certainly above 85° by feel alone.View attachment 3642791 So, here's what I'm running...
4 cxb3590 3500k (72v) on a meanwell hlg-185-700a. The sink is the 7.28 profile from heatsinksusa at a length of 24". I'm guessing that im losing all kinds of efficiency here, and possibly might damage the CoBs at full power?

Clearly, I'm going to need some fans, so would one large central fan be enough, of should I put two equidistantly placed on each bar? Any recommendations?
Click to expand...
You need to get a thermocouple or at the very least a good meat thermometer and measure directly behind the cobs and as close to them on the front as possible too. To get a real estate of what they are at.

55-60c is too hot to touch for very long by hand for most people. But in their heads think of it as "cool". It's hot by human standards...cool by led.
 
NM High Desert

NM High Desert

Active Member
Thanks PurpleBuz and Alesh
I felt better when I saw the output volts and amps.

I was starting to suspect my KillaWatt, it is new.
I tested it on a source that I new was pulling 30 watts and it read 28.
Then tested it on my t5 VHOs which are rated 385 and it read 405.
The middle range could be way off, I may have to buy another for comparison.

Thanks again.
Enjoy your Easter!
 
BOBBY_G

BOBBY_G

Well-Known Member
UniversZero said:
So, I may have grossly overestimated the passive cooling abilities of my bars. After running for an hour or so, these guys get pretty hot... Not so much that I can't keep my finger on them, but seem to be much hotter than I would have thought... Certainly above 85° by feel alone.View attachment 3642791 So, here's what I'm running...
4 cxb3590 3500k (72v) on a meanwell hlg-185-700a. The sink is the 7.28 profile from heatsinksusa at a length of 24". I'm guessing that im losing all kinds of efficiency here, and possibly might damage the CoBs at full power?

Clearly, I'm going to need some fans, so would one large central fan be enough, of should I put two equidistantly placed on each bar? Any recommendations?
Click to expand...
at 120 cm^2/heat watt you need 30" of that heatsink, you have 24, so it is a bit undersized

1 fan would prob get you there. 2 would run it cooler and prob give you a very slight bump in brightness

also why does 85 deg (im assuming you mean Fahrenheit) matter? thats actually pretty darn cool. if you can keep your finger on it continuously, its prob 120F or less (50C)
 
Last edited:
D

dastingoo

New Member
Can I use a single driver to power two bars of lights by splitting up the wires from the driver into two?

So basically, you have the driver and from that leaves a trail of + and - to both bars.
As opposed to just using one trail of +/- going from bar to bar.

This is merely for aesthetics.
 
U

UniversZero

Active Member
BOBBY_G said:
at 120 cm^2/heat watt you need 30" of that heatsink, you have 24, so it is a bit undersized

1 fan would prob get you there. 2 would run it cooler and prob give you a very slight bump in brightness

also why does 85 deg (im assuming you mean Fahrenheit) matter? thats actually pretty darn cool. if you can keep your finger on it continuously, its prob 120F or less (50C)
Click to expand...
Thanks for addressing my concerns, BG. After reading up about the whole binning process, as it turns out I meant 85°C... Which would put them well within normal operating ranges. I had repeatedly seen that nominal figure strewn across various DIY cob build threads and mistakingly thought the designation was Fahrenheit.
Good to know I'm only undersized to the tune of ~20%... It will give me the opportunity to wire a few fans and learn something new anyways.

Follow-up question though... If they are indeed running at somewhere around 50°C, what are the consequences to running these bars without fans...just decreased longevity? I'm not overly worried about slight increases in brightness.
 
U

UniversZero

Active Member
knight mare said:
Yes those, yes it needs a heat sink and a driver. but it would keep your light very cold so you could run it harder.
Click to expand...
They seem incredibly inefficient at producing the cooling effect. Have you used them in this application before, and if so, have you measured the wattage draw?
 
guod

guod

Well-Known Member
UniversZero said:
They seem incredibly inefficient at producing the cooling effect. Have you used them in this application before, and if so, have you measured the wattage draw?
Click to expand...
Thermoelectric junctions are about 4 times less efficient in refrigeration applications than conventional means (they offer around 10–15% efficiency of the ideal Carnot cyclerefrigerator, compared with 40–60% achieved by conventional compression cycle systems (reverseRankine systems using compression/expansion).[6]) Due to this lower efficiency, thermoelectric cooling is generally only used in environments where the solid state nature (no moving parts, low maintenance, compact size, and orientation insensitivity) outweighs pure efficiency.

https://en.wikipedia.org/wiki/Thermoelectric_cooling
 
U

UniversZero

Active Member
guod said:
Thermoelectric junctions are about 4 times less efficient in refrigeration applications than conventional means (they offer around 10–15% efficiency of the ideal Carnot cyclerefrigerator, compared with 40–60% achieved by conventional compression cycle systems (reverseRankine systems using compression/expansion).[6]) Due to this lower efficiency, thermoelectric cooling is generally only used in environments where the solid state nature (no moving parts, low maintenance, compact size, and orientation insensitivity) outweighs pure efficiency.

https://en.wikipedia.org/wiki/Thermoelectric_cooling
Click to expand...
How favorably would this compare with the more standard approach of fan-on-heatsink? Say I were displacing 50w of cob (so at ~56% eff, that's what... 22w of heat?)... What sort of wattage draw are we talking to run that cob acceptably cool? Will condensation issues be a factor when running these in such an application?
 
K

knight mare

Well-Known Member
I used them year's ago, in computers to cool cpu's it should allow you to use a smaller heatsink if you used a temperature probe...And an Arduino to power it and keep it at a set temperature sounds like it could be a good project to try.
 
K

knight mare

Well-Known Member
As to the condensation, that would depend on how hard you have to drive them to keep the temperature were you want it.But I do no think that it would be a problem.
 
You must log in or register to reply here.
Top