Reef Central Online Community

Go Back   Reef Central Online Community > General Interest Forums > Do It Yourself
Blogs FAQ Calendar Mark Forums Read

Notices

User Tag List

Reply
Thread Tools
Unread 10/16/2018, 07:50 PM   #1
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
calling all Alluminum heatsink expert

hi all I need some input /help

I was given 2 pieces of Aluminum plates
their measure is
L 48" x W 4 1/4 x 1/4 thick and around 4-5 pounds in weight
the question is....
can I use this as heat sink for a DIY Led Project?

thanks.

[IMG]Aluminum plates

[IMG]Aluminum plates

[IMG]Aluminum plates


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/17/2018, 02:10 AM   #2
Gorgok
Registered Member
 
Gorgok's Avatar
 
Join Date: Feb 2011
Location: Cypress, Texas
Posts: 1,904
Short answer? Yes, maybe, and no.

Longer answer is it depends. How much heat you are pumping into it, for how long, in what sort of conditions, etc...

It won't be good at getting rid of heat (relatively speaking low surface area, and its thick) but it will soak some up for a while. That while could be enough, or it could be lots too little...

How much heat are you planning on making?


Gorgok is offline   Reply With Quote
Unread 10/17/2018, 04:44 AM   #3
mcgyvr
Registered Member
 
mcgyvr's Avatar
 
Join Date: Sep 2003
Location: North Carolina
Posts: 20,050
^ yep..

Of course it can be a heatsink...
Most black box LEDs just use an aluminum plate that has no fins at all..
If its sufficient or not for your build depends on the specifics of your build..
It will easily handle all of those LEDs in the back of picture 1...

Being a "good" heatsink is about surface area.. A flat plate by itself doesn't have nearly the surface area of one with tons of fins/pins or whatever...


__________________
Who me?
mcgyvr is offline   Reply With Quote
Unread 10/17/2018, 05:46 AM   #4
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
Hi guys thanks for the reply
I'm planning to put 12 leds of 10 watts each ,on each plate
Thanks again


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/17/2018, 06:12 PM   #5
oreo57
Registered Member
 
Join Date: Apr 2013
Posts: 1,793
no expert and thermal management is sort of part art,science, and trial ..BUT this is a bit of a "rule of thumb" approach..

Quote:
If you want the LEDs to run a long time and be the most efficient, then 12 sq in per watt is a good "conservative" place to work. This will work for pretty much any LEd.

The other end of the thermal area needed - about 1 in sq per watt - is on the aggressive side, and will result in higher die temperatures, somewhat shorter life, and slightly less efficient.
http://www.candlepowerforums.com/vb/...for-power-leds
you have approx 408 sq inches...thickness is really of lesser importance and adds little, at about "2W" of heat dissipation w/ the conservative approach..


So one "rule" says you can do like 430W another one says 35W...

Pick one or in-between...

Now that also assumes you are actually going to RUN them at 10W which, if cheap chips, wouldn't be recommended anyways..


oreo57 is offline   Reply With Quote
Unread 10/17/2018, 08:03 PM   #6
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
Wow didn't know
All these rules
And yes they are two
Opposite data's
12 sqi x watt ??
Wondering what the
Companies are using
For the black box


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/17/2018, 08:08 PM   #7
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
Thanks Oreo
Definitely sucks though
I have drivers that have to run
12 at the time


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/18/2018, 10:17 AM   #8
mcgyvr
Registered Member
 
mcgyvr's Avatar
 
Join Date: Sep 2003
Location: North Carolina
Posts: 20,050
12in^2 per watt? ha ha ha ha...
Black box LEDs are doing WAY more wattage than that...


__________________
Who me?
mcgyvr is offline   Reply With Quote
Unread 10/18/2018, 11:33 AM   #9
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
Quote:
Originally Posted by mcgyvr View Post
12in^2 per watt? ha ha ha ha...
Black box LEDs are doing WAY more wattage than that...
Hi mcgiver
Can you clarify please. As I'm not understanding...
Thanks


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/18/2018, 01:10 PM   #10
oreo57
Registered Member
 
Join Date: Apr 2013
Posts: 1,793
Quote:
Originally Posted by mcgyvr View Post
12in^2 per watt? ha ha ha ha...
Black box LEDs are doing WAY more wattage than that...

first off I want to say 12 sq inches per watt IS extreme but ssumes passive cooling..

BB's use 2 fans....and run about 2W at the diodes..
100W-ish is probably a more realistic "guess"..
180 sq " boards..Lets just say 2 sq " per watt w/ active cooling..


Reefbreeders use finned heatsinks and fans on some of their boxes..

Actually surface area "could be" is much larger.. based on this:
It's a Mars Aqua hort light. much bigger than a black box though.


CORRECTIONS:
suspect smaller bb's have no finned heat sink
Many use 3 fans for cooling..

@12square inches per real watt. for 100w flat plate would be around 34"x17" plate passive.. w a large margin of error (not really subtracting any diode footprints.)



Last edited by oreo57; 10/18/2018 at 01:45 PM.
oreo57 is offline   Reply With Quote
Unread 10/18/2018, 02:00 PM   #11
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
So if I install 24 leds of 10 watts each, and run them at 900 MA
or 850 ma instead of 1000 and also installing 3 fan on the top should be ok? Not the best but ok?
Thanks


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/18/2018, 09:34 PM   #12
oreo57
Registered Member
 
Join Date: Apr 2013
Posts: 1,793
Personally, I dislike fans but a lot depends on where you put the lights.. You encoded? Open top ect..

I really don't think you will have an issue w/out the fans and lower currents...900-ish might push each to it's 10W depends on the chip..

to be honest, it's going to be a bit of trial and error..

What drivers are you planning on using?


oreo57 is offline   Reply With Quote
Unread 10/21/2018, 06:35 AM   #13
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
Hi there I have 2 meanwell
Hlg 120 H I have using one for 3 year now


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/21/2018, 09:56 AM   #14
oreo57
Registered Member
 
Join Date: Apr 2013
Posts: 1,793
There are dozen of models of HLG120's..


anyways I guess you are planning on running the 10W chips in parallel to divide the driver output current..

Since most common 10W chips run around 34V V(f) and 120 drivers capable of that voltage have 4A or less available..

Soo @ 6 per driver your max current to each is 600mA to 300mA


oreo57 is offline   Reply With Quote
Unread 10/21/2018, 12:23 PM   #15
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
Hi there the leds are in Serial,
The driver puts out
120V at 1a
Right now on my old rail I'm running 11 leds
The leds can be run at 10 to 12 volts
120 /11 = 10.9 volts @1000 mA


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/25/2018, 08:33 PM   #16
OzIA
Registered Member
 
Join Date: Sep 2016
Posts: 214
I suggest you run some tests if you already have the LEDs and drivers. Take some readings and see where the temp starts to level out. Surface area is the key but a small fan can make a big difference.

Typically, LED intensity starts to diminish above 105 deg f and if you run them long over 135, they won't last.

I've been using these 3.5 inch pucks that you can find on eBay. They have less than 10 sq in on each but there are (15) one watt LEDs on them, so.. far greater than the amount recommended above. There are no fins but i do have a small fan blowing on four of them. Without the fan they would burn up in a day or two.

I put a temp probe on one of them and monitor it with the Apex. These are my refugium lights. If the temp goes above 110, I shut them off.

I've attached the temp graph.


Attached Images
File Type: jpg screenshot_109.jpg (45.3 KB, 9 views)
OzIA is offline   Reply With Quote
Unread 10/25/2018, 09:13 PM   #17
Scrubber_steve
I'm really very likeable
 
Scrubber_steve's Avatar
 
Join Date: Aug 2016
Location: Down Under
Posts: 676
I can't see that alloy plate being useful much at all. The life of the LEDs will be diminished greatly if they are not cooled efficiently.

Basically, an LED produces a lot of heat behind the chip & this heat needs to be sucked, or absorbed away from the chip to keep the chip at a reasonable operating temperature.
A heatsink absorbs this heat. But the heatsink must be able to cool quickly & efficiently itself so it doesn't over heat. If it can't cool rapidly enough its temperature gets closer to the temperature produced at the chip & its ability to absorb enough heat away from the chip is drastically diminished. So the heatsink cannot be made from a material (normally aluminum) that is too thick because it will hold the heat. As already pointed out fins greatly increase the surface area of a heatsink and that enables it to radiate heat - loose heat, much more quickly/efficiently.

This is a heatsink I use. The base is only 2mm thick, & the fins 1mm thick.


Attached Images
File Type: jpg heatsink.jpg (40.6 KB, 19 views)
Scrubber_steve is offline   Reply With Quote
Unread 10/25/2018, 09:28 PM   #18
Isayso
Registered Member
 
Isayso's Avatar
 
Join Date: Oct 2003
Location: Hamilton,ON.Canada
Posts: 527
Thanks ozla
I need to check the temp
I think tha is above the 100 f

Thanks for the suggestion Twinfallz
Did you buy that heatsink locally?


__________________
lets save the reef

Current Tank Info: Hardware: 125g Reef,Mak4 for return,DIY 20G sump,DIY Beckett Skimmer(AF30) driven by a Mak4,DIY S.Denitrator,DIY Calcium reactor(PM), wavemaker .MJ mods on Swirlestein,Jebao WP 20,Lighting: 60 LEDs, moon lights
Isayso is offline   Reply With Quote
Unread 10/25/2018, 09:37 PM   #19
Scrubber_steve
I'm really very likeable
 
Scrubber_steve's Avatar
 
Join Date: Aug 2016
Location: Down Under
Posts: 676
Quote:
Originally Posted by Isayso View Post

Thanks for the suggestion Twinfallz
Did you buy that heatsink locally?
No, I had that extruded to my own design; but you can buy similar on ebay or aliexpress


Scrubber_steve is offline   Reply With Quote
Unread 10/26/2018, 11:24 PM   #20
oreo57
Registered Member
 
Join Date: Apr 2013
Posts: 1,793
Quote:
Originally Posted by Isayso View Post
Thanks ozla
I need to check the temp
I think tha is above the 100 f

Thanks for the suggestion Twinfallz
Did you buy that heatsink locally?
FIRST I'm not arguing against "proper heatsinks" but this isn't exactly what you asked about.
This is enough to passive cool 10W chips (at least for me for a few years b4 abandoning the build). Granted a lot of surface area and a lot of convection air movement arounfd the sink but not really very big.
Temps would easily go over 120F..

SECOND there is a large difference between cheap chips and quality modern chips:
Quote:
The XB-D LED, among many other new
Cree XLamp LEDs, is binned at 85 °C, so the relative flux data is based on 100% light output at an 85 °C junction temperature.
That's 185F "at the core"...cheaper chips should probably be kept below 158f..
NOTE: That is at the core not the sink.


Easiest "fix" if you think the chips are heating too much is just tweak the drive current down a bit..
Less current less heat but oddly the LED's become more efficient..

Oh and yes black anodizing.. and even black paint radiates heat off the sink better than a raw or polished surface. Not needed for fan cooled though.

I've seen plenty of things that shouldn't work, work just fine..

Would it be my choice? No. But it's free, large and, if needed, can be enhanced w/ thermal pasting some fins on top..

Cheap 10W chips are like $2.50 each...

Again, heat sinking is complicated and even manf need to do Real world tests over any calcs and guessing..



Last edited by oreo57; 10/26/2018 at 11:39 PM.
oreo57 is offline   Reply With Quote
Unread 10/27/2018, 01:50 AM   #21
Scrubber_steve
I'm really very likeable
 
Scrubber_steve's Avatar
 
Join Date: Aug 2016
Location: Down Under
Posts: 676
The color of a heatsink plays no part in its ability to radiate heat (loose heat - cool down). A coating of paint will act as an insulation & restrict heat loss in fact. A black surface absorbs light energy better than another colored surface, though this has nothing to do with heatsinks cooling electronic components.


Scrubber_steve is offline   Reply With Quote
Unread 10/27/2018, 09:49 AM   #22
oreo57
Registered Member
 
Join Date: Apr 2013
Posts: 1,793
Quote:
Originally Posted by Twinfallz View Post
The color of a heatsink plays no part in its ability to radiate heat
I disagree and so does science..


Quote:
In situations where convection is low, such as a flat non-finned panel with low airflow, radiative cooling can be a significant factor. Here the surface properties may be an important design factor. Matte-black surfaces will radiate much more efficiently than shiny bare metal.[17][18] A shiny metal surface has low emissivity. The emissivity of a material is tremendously frequency dependent, and is related to absorptivity (of which shiny metal surfaces have very little). For most materials, the emissivity in the visible spectrum is similar to the emissivity in the infrared spectrum[citation needed]; however there are exceptions, notably certain metal oxides that are used as "selective surfaces".
Quote:
Also,
the effect of radiation heat transfer is
very important in natural convection, as it can be responsible of up to 25% of the total heat dissipation. Unless the
component is facing a hotter surface nearby, it is imperative to
have the heat sink surfaces painted or anodized to enhance radiation.
https://www.boydcorp.com/files/techn...-Heat-Sink.pdf


OR:
Quote:
Effects of Anodising on radiant heat transfer in heat sink

Heat sinks cooled by natural convection may benefit from an anodized finish, but the added cost of the finish may not be justified when the part is used in forced convection cooling.

Surface emissivity limits the amount of heat transfer due to radiation cooling. With 1.0 being perfect (black body) emissivity anodized aluminum is 0.85 and unfinished is 0.05.

Heat transfer due to radiation is proportional to the heat sink surface area exposed to its surroundings and to the temperature rise above ambient (in absolute °K) raised to the 4th power (T Heat-sink - Tambient)4. In natural convection on small heat sinks with open fins, and a high benefit from anodizing by up to 45%.

Relatively large extrusions and those used at low temperature rise, as in many high power applications, will only gain up to 10% by the addition of an anodized surface.

With forced ventilation (using a fan) convective cooling is about 3 times higher than in natural convection. This changes the proportion of heat transfer due to radiation. An anodized finish will only add 4 -8% to the overall cooling effect in forced air. This percentage again, depends on fin spacing and heat sink dimensions. The color of the anodized finish makes little impact on emissivity since most radiational heat loss occurs at wavelengths higher than visible light.

As a thumb rule, if anodize is not required for aesthetic or corrosion protection, we suggest it only for small, open finned heat sinks in natural convection.
http://smlease.com/entries/heat-tran...r-in-heat-sink

Like I said, w/ fans.. not an issue..



Last edited by oreo57; 10/27/2018 at 10:19 AM.
oreo57 is offline   Reply With Quote
Unread 10/27/2018, 03:11 PM   #23
Scrubber_steve
I'm really very likeable
 
Scrubber_steve's Avatar
 
Join Date: Aug 2016
Location: Down Under
Posts: 676
Oreo57, let’s take a closer look at the info you posted.

Firstly, - “Matte-black surfaces will radiate much more efficiently than shiny bare metal.”
Why?
You then quoted, - “A shiny metal surface has low effective emissivity”. But areo57, you cut that quoted sentence short.
The full sentence states, - “A shiny metal surface has low effective emissivity due to its low surface area.
Yes, a flat black surface is just that because it rough, it has greater surface area & absorbs more light. Whereas a shiny surface is shiny and reflects a lot of light because it’s smoother – less surface area. The greater the surface area the greater the radiative cooling (i.e. fins).

Also think of it this way, -how shiny is that metal surface in pitch blackness? An object has no color in the dark. In the dark how could color possibly have any effect on its radiative ability? Surface area does affect radiative ability, in light or darkness.

Your next quote states, - “Unless the component is facing a hotter surface nearby, it is imperative to
have the heat sink surfaces painted or anodized to enhance radiation.”

The author neither states why or how “painting” or “anodizing” enhances radiation, or the color the paint or anodizing should be. Black is not mentioned.


Which brings us to your last quote. – “Heat sinks cooled by natural convection may benefit from an anodized finish”

But, further along in the article it is stated, & I quote “The color of the anodized finish makes little impact on emissivity since most radiational heat loss occurs at wavelengths higher than visible light."

.


Scrubber_steve is offline   Reply With Quote
Unread 10/27/2018, 09:20 PM   #24
oreo57
Registered Member
 
Join Date: Apr 2013
Posts: 1,793
Believe what you want to believe.. Black Anodizing increases passive cooling.

Just a known fact..

Yes other colors help too.

If you can find ANY facts that state black anodizing "hurts" heat transfer under our conditions I'd be REALLY surprised....

Quote:
4.8. Black surfaces
Under natural convection conditions, the performance of a heatsink with a black surface will be 6% to 8% better than that with a plain or bright surface. However, this differential disappears under forced air conditions.
Yes, it's complicated but I stand by my general pricipals in regard to passive cooling..
http://robots.freehostia.com/Heatsinks/Heatsinks.html


https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5455404/

Quote:
To test the influence of thermal radiation on the total heat transfer in buoyancy-induced convection, an untreated and painted/anodized version of each heat sink was tested. Both painting and anodizing increase the emissivity of the heat sink and, consequently, the heat transfer by thermal radiation......Figure 8 compares the thermal performance of the untreated finned heat sink, the 22 mm-high in-house-made aluminum foam heat sink (reticulated22) and the 22 mm-high solid aluminum block to that of their painted/anodized counterparts. By painting or anodizing, the thermal performance improved for each of the tested heat sinks. The aluminum foam heat sink, the finned heat sink and the aluminum block, respectively, dissipated on average 17%, 31% and 44% more heat after they were painted or anodized.




Last edited by oreo57; 10/27/2018 at 09:46 PM.
oreo57 is offline   Reply With Quote
Unread 10/27/2018, 09:48 PM   #25
Scrubber_steve
I'm really very likeable
 
Scrubber_steve's Avatar
 
Join Date: Aug 2016
Location: Down Under
Posts: 676
Quote:
Originally Posted by oreo57 View Post
Believe what you want to believe.. Black Anodizing increases passive cooling.

Just a known fact..
Just to quote again from the article you supplied - “The color of the anodized finish makes little impact on emissivity since most radiational heat loss occurs at wavelengths higher than visible light."

so again I'll ask; how could the actual color of a heatsink affect its ability to loose heat when operating in an environment of near dark, to, & especially, total darkness? It is light that gives an object its color; the color spectrum that an object cannot absorb, but reflects is the color it appears to the human eye. So in near, or total darkness a heatsink has no color.
A black surface has a greater surface area than other colors & absorbs more light.

Quote:
Originally Posted by oreo57 View Post
If you can find ANY facts that state black anodizing "hurts" heat transfer under our conditions I'd be REALLY surprised....
I never suggested that. But painting a heatsink is never recommend because the paint, especially the thicker the coat, will act as an insulator & inhibit the heatsink from cooling.


Scrubber_steve is offline   Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On



All times are GMT -6. The time now is 02:26 AM.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Powered by Searchlight © 2024 Axivo Inc.
Use of this web site is subject to the terms and conditions described in the user agreement.
Reef CentralTM Reef Central, LLC. Copyright ©1999-2022
User Alert System provided by Advanced User Tagging v3.3.0 (Pro) - vBulletin Mods & Addons Copyright © 2024 DragonByte Technologies Ltd.