Heatsinks for DIY LED lamps

[AquariusPanta]

Increased power again, dissipating 108W, same heatsink, now 50cm²/dissipation W

Ambient temp 22.5C
Heatsink temp reached 40C with no airflow, convection only
light loss was 7% versus ambient baseline

Added 120mm fan running softly at 5V (dissipating .463W) consuming .555W.
Heatsink temp stabilized at 28C
light loss was 3.36% versus ambient baseline

I increased fan power to 9.35V (dissipating 1.56W) consuming 1.87W
heatsink temp stabilized at 26.5C
light loss was 2.94% versus ambient baseline

So when you work the numbers we get:
baseline efficiency 42%
Passive cooling 39%
Active cooled 5V = 40.4%
Active cooled 9V = 40.1%

Because of the white COB's surprising performance with heat, the active cooling had less benefit than expected and once again the 5V was actually better than 9V. This test is important so I intend to repeat it with the 5.88" profile next time. I have a 12" length with a 140mmm fan and a pair of CXA3070 ABs mounted.

I imagine replacing the Cree 3070 ABs with a different COBs series, such as the Vero 18 or 29, may or may not require either more cooling or more aluminum, depending on known values of thermal resistance.

With the 5V/9V testing, are you simply supplying the fan(s) with a 5V and 9V PSU? If so, what amperage are they ticking at? I've had bad luck with ordering PSUs for fans (Intel Fan/HeatSink), with stated currents NOT matching operational currents (fans not spinning completely).

Also, are you performing the tests in a tent or a room, such as a basement? What are the dimensions? I find this all to be relative in best understanding your results.

 

[AquariusPanta]

The stock cooler will work great for almost every COB, look up the heat removal specs, expressed at TDW i think, usually up around 90 watts


CPU stock coolers are rated at TDW, usually 90 watts for power (heat) displacement is acheivable

max. core temps for some CPU

I have a Sandy Bridge. Does that mean it starts throttling at 69-72C? (assuming it even gets that hot)

And you're positive that's what those charts are telling you?

I think the temperatures on those charts are only max ratings, not threshold settings. So what type of bios do I have? What is my CPU frequency set to ? what is the voltage set to? Am I using a laptop? Do I even have speedstep activated?

I had bought a very large aftermarket heatsink to overclock this CPU in 2012, but got negligible frequency gains with sandy bridge, so I put it back to stock. I'm not using the stock cooler.

...Nerds...

 

[SupraSPL]

(Enter captain caveman) so if I'm understanding this right, you can get away with running 2 cobs at 60.5 watts each passively without a fan, per heat sink?
If it's only 1% less efficient than mounting each one to say an arctic11, I would much rather go the heat sink route even if it's a little more in initial cost.
More streamlined and quieter.
Thanks for running these tests Supra!

That was 60.5W total on the heatsink, but I did try 108W on that same heatsink and the light loss was 7%. So in this case adding a slight breeze was a huge help. It may be that you circulation fan will add enough of a breeze to get the same effect.

 

[SupraSPL]

I haven't had the same good results as you running low fan speeds. That heatsink must work especially well with cobs. Anytime I slow my fan speeds the heat builds up quickly..but my ventilation is very low cfm

The low fan speeds do allow the heatsink temp to rise a bit, but I was very surprised and the COBs ability to maintain its output at those higher temps. It kind of mocks my attempts to achieve a low Tj So I am realizing if I want to increase gr/W, it is going to be more about spreading the light as evenly as possible into the canopy and running as low of a current as I can afford with CXAs.

 

[SupraSPL]

Do you have the capability of running the system on the 10.08 profile at around 180watts, or 30cm^2/w? I'm wondering what the upper limit would be with passive cooling.

Do you know if there would be a difference in efficiency if you used pwm to dim the fan, as opposed to adjusting the voltage?

Thanks for testing this stuff out for us. Saves a lot of trial and error. By the way, how did you use the superscript on this forum?

Unfortunately I only have 1 2.3A driver on hand but I have another one on the way so I could test it then. When I test the 5.88" profile I will run it at 30cm²/W passive and see if the temp stabilizes. I did something similar when I was testing the CPU coolers with no fan running. I let them get super hot (heatsinks 50C) and saw as much as 22% light loss with the Vero 18 97 cri.

Good question about using PWM to dim the fan. For those who use automated systems I imagine that could be very handy and if you have a high efficiency PSU, won't waste power. But for simple designs I suppose it would be most efficient to use power adapters that are the voltage you want so they do not need to be dimmed. You can get 3V 4.5V 5V 6V 7.5V 9V 12v and 15V oem adapters on eBay relatively cheaply. If you run them at least half of their current, they tend to run at 80% efficiency. Look for the "efficiency level V" designation but 3 and 4 are good as well.

Superscript is alt + 0178. It only seems to work on my number pad though.

 

[SupraSPL]

44 inches is 111.76 cm.111.76 x 84.43=9435.89 cm 2
that is the formula right?

I calculate the perimeter of the 2.08" profile to be 13.09" or 33.25cm. So it is 111.76cm X 33.25cm = 3716cm².

 

[SupraSPL]

I imagine replacing the Cree 3070 ABs with a different COBs series, such as the Vero 18 or 29, may or may not require either more cooling or more aluminum, depending on known values of thermal resistance.

With the 5V/9V testing, are you simply supplying the fan(s) with a 5V and 9V PSU? If so, what amperage are they ticking at? I've had bad luck with ordering PSUs for fans (Intel Fan/HeatSink), with stated currents NOT matching operational currents (fans not spinning completely).

Also, are you performing the tests in a tent or a room, such as a basement? What are the dimensions? I find this all to be relative in best understanding your results.

That is true, if you had a certain threshold for light loss, and the Veros have a better thermal resistance, as long as they do not suffer from a greater temp droop, you could use less cooling to achive the same level of light loss. I plan to get a Vero 29 to test this out. CXA3070s and Veros have different strengths. I think the Vero would be awesome for a commercial lamp run hard but the CXA is best suited for DIYers to run soft.

Yes I am just switching between a 5V and 9V adapter powering the same 120mm fan. With 5V adapter it measures 4.93V * .094A (dissipating .463W) consuming .555W assuming 20% loss in the adapter. With 9V adapter it measures 9.35V * .167A (dissipating 1.56W) consuming 1.87W assuming 20% loss in the adapter. A good compromise in this case might be a 7.5V adapter, but it depends on the fan you are working with. I recommend avoiding non-oem adapters.

The tests are performed in a large room with steady air temp and very minimal airflow. The most important thing is that the light source and light measuring device are not moved even slightly when swicthing fan speeds. I place the lights at whatever distance puts the meter at about 50% of it range (200,000 lux meter). I do my best to make sure the drivers are warmed up before the test so their output does not change too much as the COBs warm up. I use a baby food infrared thermometer to monitor heatsink temps (between fins). I am using a cheap lux meter so there is no way I can accurately quantify the light, the meter has poor accuracy but very good precision and consistency. So the tests are qualitative only, which is all we need in this case. So it is a simple setup and should be easy for anyone to replicate these tests.

 

[AquariusPanta]

The tests are performed in a large room with steady air temp and very minimal airflow. The most important thing is that the light source and light measuring device are not moved even slightly when swicthing fan speeds. I place the lights at whatever distance puts the meter at about 50% of it range (200,000 lux meter). I do my best to make sure the drivers are warmed up before the test so their output does not change too much as the COBs warm up. I use a baby food infrared thermometer to monitor heatsink temps (between fins). I am using a cheap lux meter so there is no way I can accurately quantify the light, the meter has poor accuracy but very good precision and consistency. So the tests are qualitative only, which is all we need in this case. So it is a simple setup and should be easy for anyone to replicate these tests.

I wonder how dramatically different these tests would be if held in a medium sized grow tent. My small tent (5' x 3' x 1.7') would get super hot (upper 80's) in my closet (9' x 2' x 10') if I didn't have the closet doors wide open during operation, with ambient (surrounding environment) temperatures running around 70 degrees fahrenheit.

This of course is affected by how many LEDs are running and incoming/outgoing air flow, so many tests will be needed just in order to accurately generalize when considering passive or actively cooling. I guess us diehards will have to, over time, accumulate such data over time to establish some common consensus.

 

[bicit]

I wonder how dramatically different these tests would be if held in a medium sized grow tent. My small tent (5' x 3' x 1.7') would get super hot (upper 80's) in my closet (9' x 2' x 10') if I didn't have the closet doors wide open during operation, with ambient (surrounding environment) temperatures running around 70 degrees fahrenheit.

This of course is affected by how many LEDs are running and incoming/outgoing air flow, so many tests will be needed just in order to accurately generalize when considering passive or actively cooling. I guess us diehards will have to, over time, accumulate such data over time to establish some common consensus.

Keep in mind that there would be no difference in ambient room temps between a passive or actively cooled system since they operate at about the same level of efficiency. Just because the heat sink is hotter does not mean that more heat is being created.

 

[Greengenes707]

@AquariusPanta... I don't see passive cooling being the best option in a hot ambient environment. If ambient is already 85*+...it won't take much to get into temps that start to effect longevity, not just output at that second in time. If ambient is borderline to start with...actively cooling is the best way to keep temps as low as the environment will allow.
@bicit... If the heatsink is enough hotter(horrible grammar)...it will effect ambient faster than a cooler one...even if they technically "dissipate" the same amount of heat. Same concept that make HPS so hot. A 600 hps is 42% efficient...but I wouldn't go grabbing on to one.
@SupraSPL... Thank you so much for the time spent to conduct these tests. Very valuable stuff here. I'm excited to see how the 5.8 profile performs.

 

[bicit]

@bicit... If the heatsink is enough hotter(horrible grammar)...it will effect ambient faster than a cooler one...even if they technically "dissipate" the same amount of heat. Same concept that make HPS so hot. A 600 hps is 42% efficient...but I wouldn't go grabbing on to one.

No, that is not the case. The only thing a fan does for an actively cooled light is move the heat which gives the (subjective) perception of 'lower temperatures' since the energy is dispersed throughout the volume of the space more evenly. With a passively cooled unit with mechanical air movement, the energy would be more concentrated in the space above the heatsink, while the space below the heatsink would remain cooler. The total energy displaced though, would be the same, thus T final would be the same without something to absorb and store some of the energy(plants), or airflow exchange.

Most of the power used to grow a plant is ultimately released as heat to the environment(practically all of it), it's source is irrelevant. A 600 watt HID will add the same amount of heat to a space as a 600 watt LED or even a 600 watt space heater. However nothing would grow in the space heater space.... Maybe a potato.

 

[SupraSPL]

True, the results in use will be different than the tests because my test room ambient is 21C but the ambient in the grow tent with lights on (should be) 25-26C. So these percentages and CM2 numbers are not quantitative either, this is just to give us some understanding of how the temps are affected by airflow and how the COBs respond to the temps. Maybe I can rig something up to run the test in the actual tent so we get a better lock on how much cm² we need.

 

[Dloomis514]

http://usefulshortcuts.com/downloads/ALT-Codes.pdf

Hold down the "Alt" key then enter the code on the numeric keypad with Num Lock on (the number keys at the top of the keyboard wont work) And if the string has a zero in it, the zero is definitely needed

Unfortunately I only have 1 2.3A driver on hand but I have another one on the way so I could test it then. When I test the 5.88" profile I will run it at 30cm²/W passive and see if the temp stabilizes. I did something similar when I was testing the CPU coolers with no fan running. I let them get super hot (heatsinks 50C) and saw as much as 22% light loss with the Vero 18 97 cri.

Good question about using PWM to dim the fan. For those who use automated systems I imagine that could be very handy and if you have a high efficiency PSU, won't waste power. But for simple designs I suppose it would be most efficient to use power adapters that are the voltage you want so they do not need to be dimmed. You can get 3V 4.5V 5V 6V 7.5V 9V 12v and 15V oem adapters on eBay relatively cheaply. If you run them at least half of their current, they tend to run at 80% efficiency. Look for the "efficiency level V" designation but 3 and 4 are good as well.

Superscript is alt + 0178. It only seems to work on my number pad though.

 

[Greengenes707]

No, that is not the case. The only thing a fan does for an actively cooled light is move the heat which gives the (subjective) perception of 'lower temperatures' since the energy is dispersed throughout the volume of the space more evenly. With a passively cooled unit with mechanical air movement, the energy would be more concentrated in the space above the heatsink, while the space below the heatsink would remain cooler. The total energy displaced though, would be the same, thus T final would be the same without something to absorb and store some of the energy(plants), or airflow exchange.

Most of the power used to grow a plant is ultimately released as heat to the environment(practically all of it), it's source is irrelevant. A 600 watt HID will add the same amount of heat to a space as a 600 watt LED or even a 600 watt space heater. However nothing would grow in the space heater space.... Maybe a potato.

I know the theory of BTU and heat output...but it doesn't translate to real life. That is the problem with much of this theory talk we get into. Also notice the term "faster" that I used...important there.
Without the heat being moved away form the source, it will add up and become hotter in the environment...I have tested it. Which is the reason I would leave an at600 in a tent....but not 650w space heater.

 

[guod]

shortcuts...
[Alt Gr] + 2 => ²
[Alt Gr] + 3 => ³
[Alt Gr] + m => µ

[CTRL]+[ALT]+ 2 => ²
[CTRL]+[ALT]+ 3 => ³
[CTRL]+[ALT]+ m => µ
...

 

[AquariusPanta]

Keep in mind that there would be no difference in ambient room temps between a passive or actively cooled system since they operate at about the same level of efficiency. Just because the heat sink is hotter does not mean that more heat is being created.

The point I may have not underlined or that may be too obvious for some, including yourself, was that if you grow in a larger room in comparison to a tent within a closet, the larger room is going to better handle (@Greengenes707 , hehe) larger amounts of heat than the inside of the tent, given ambient temperatures are around the same.

For example (real life), with 2x Vero 18's (60W combined draw) and a BlackStar V2 (140W draw) operating, I can see tent (5' x 3' x 1.7') temps (@Canopy) rise to around 82 degrees fahrenheit.

Variables worth mentioning:

6" Air King constantly pushing in fresh, quasi-ambient air into tent (on low, 10W draw),

4" Inline Fan+Filter Combo sucking out hot air from the top of the tent outside (set on medium, 35W draw) every other 15 minutes,

Closet doors wide open to allow 12' x 12' x 10' room to exchange heat.

If I close the closet doors, the tent temperatures skyrocket in no time, into the lower 90's, even with the fans running.
When the lights are turned off and an hour or so has passed by, the tent temperatures fall to around 67 degrees fahrenheit (similar to ambient temperatures).
That fucking BlackStar emits some serious heat though.

What you say though Bicit, is questionable. While I agree the efficiency would be similar, the temperature of the actively cooled HeatSink would be lower in temperature in comparison to the passively cooled HeatSink, according to Supra's test results. This means that while the amount of heat emitted between the two methods would quantitatively be similar, the temperature of the the heat from the passively cooled HeatSink would be much higher.

A way of looking at this would be like this:

A given COB running at 41% active and an identical COB running at 40.5% passive. They emit around 60W each (both heat and light). The former may be running temperatures of 30 degrees celsius, while the latter runs around 45 degrees celsius. Both are acceptable temperatures but you may favor the cooler running one, as ~36W of the heat generated would be cooler in temperature (this is open for debate ).

@SupraSPL
While it may not be necessary in your large room, measuring the ambient temperature rise with each method may be worth noting, especially in a grow tent.

 

[Dloomis514]

Something is amiss with the cooler/hotter analysis above. Even though one light is running at 30 and the other at 45, the heat load is the same, in the confines of the grow area. My logic tells me the heat from the actively cooled unit is very well mixed with the ambient air. The passive unit puts out the same heat, but the heat isn't mixed as well in the immediate area of the cob. 60 watts of heat is 60 watts heat, doesn't matter how the btu' s are distributed.

 

[AquariusPanta]

Something is amiss with the cooler/hotter analysis above. Even though one light is running at 30 and the other at 45, the heat load is the same, in the confines of the grow area. My logic tells me the heat from the actively cooled unit is very well mixed with the ambient air. The passive unit puts out the same heat, but the heat isn't mixed as well in the immediate area of the cob. 60 watts of heat is 60 watts heat, doesn't matter how the btu' s are distributed.

While I don't disagree with your claim that 60 watts of heat is 60 watts of heat, i.e. 1 = 1, I do, however, disagree that the heat load is the same between both examples.

If one COB is more efficient then the other, at a cooler temperature, then wouldn't it produce more light and less heat? While the thermal efficiency differences between each method may be miniscule in a larger room, they may indeed be more of a concern in smaller spaces, such as tents or closets where continuous air flow and periodic air exhaust are necessary for temperatures to be ideal.

I agree with you about the actively cooled unit dispersing the warmer air more readily, with the passive allowing heat to buildup around the COB and ultimately the HeatSink.

 

[Dloomis514]

The difference between 40.5 and 41% efficiency doesn't sound like much heat, a few watts maybe. Certainly not enough to explain such a large temperature difference

 

[PurpleBuz]

Cooling a large grow passively is near ridiculous in the amount of aluminum it takes. And every passively cooled light runs a good amount higher Tc than it's actively cooled counter part. Even active cooling a higher wattage light is a heavy project.

exactly what I have been thinking. Its a logistical nightmare in any decent sized grow to run a whole roof of aluminum. It gets in the way of everything including my noggin. All the successfull large 100% passive designs I have seen end up using the grow room ventilation to cool the lights.