Heatsink USA Calculator

xX_BHMC_Xx

xX_BHMC_Xx

Well-Known Member
Hey RIU,

I stole Supra's HSUSA table format and made it a bit... sexier. Also, added some functionality to help those of us that use this style of heat sink. This calculator should help you decide what profile to pick if you are considering this route. I'm no expert, and I would love someone with more knowledge to chime in on how the correction tables on the third page affect these calculations. Feel free to download and change it as you see fit. Open source yo.

Link: https://drive.google.com/file/d/0B4KikJkcxiy5QXZicmtWMXg4OW8/view?usp=sharing
 
Organic Miner

Organic Miner

Well-Known Member
Nice update. It would be great to add pin/radial heat sinks to this table since many DYIers are using them nowadays.
 
xX_BHMC_Xx

xX_BHMC_Xx

Well-Known Member
Oh, I should probably mention that all you have to do is enter your numbers into the white fields and the calc will give you all kinds of useful info. Not just heat sink stats, but some useful electrical info as well.
 
NoFucks2Give

NoFucks2Give

Well-Known Member
Yesterday I attempted to calculate the watts dissipated (heat flow) per inch length of the 0.601" and 1.00" HSUSA heat sinks.

First off the problem with these heat sinks is they are not anodized. This is a huge problem for a heat sink. They do not work well (at all) if not anodized.

A new shinny unanodized heatsink has an emissivity coefficient of 0.04 (may as well be zero) where one that is anodized can have up to 0.95.
The perfect black body coefficient is 1.0.
This site seems to have fairly accurate ranges for emissivity coefficients: http://www.optotherm.com/emiss-table.htm

Also these natural convection formulas are for heat sinks with the fins pointing outwards and not upwards as the top and bottom areas between the fins needs to be open to gravity. Cooling depends on the density and buoyancy of the air.


The following were calculated using a web app done by Microelectronics Heat Transfer Laboratory, University of Waterloo, Ontario, Canada
In the attached ZIP file there is an html page filled out with the 0.601" parameters. Not a user friendly app. If there is an error you get no answer. No error code, nothing.
It's easier if you start with one filled out. But you can try with a blank form too: http://mhtlab.uwaterloo.ca/cgi-bin/rect_hs

It would be nice to find two heatsink calculators that match. Here's one with different results: http://www.coolingzone.com/flash.php?id=4


0.601", 1.2 Watts heat flow per inch aluminum06inchWatts.jpg


1.00", 1.98 Watts heat flow per inchaluminum1inchWatts.jpg
 

Attachments

Last edited:
Rahz

Rahz

Well-Known Member
Anodizing only provides a small bump in performance when cooled via forced convection. I was achieving 10C over ambient using the 5.886 profile with fans, about 4.15 watts of heat per inch.

BTW, I have a good amount of the 5.886 profile sitting around, some drilled for 3590s, some drilled for Vero29 and some bare. Willing to let them go for less than half price... maybe 75 cents per inch plus shipping... usually $15-25 via Fed-Ex Ground depending on location.

6 pieces 18" drilled and tapped for 2x CXB3590 (recommend 75 watts each with a 120mm fan in the middle)
1 piece 12" drilled and tapped for 1x CXB3590
2 pieces 12" bare surface
2 pieces 30" bare surface
4 pieces 30" drilled and tapped for 4x Vero29 (I was running 45 watts per cob on these, good match for an HLG-185H-C700 with a couple fans)

Any of them could easily be re tapped for a different chip.

The ones that have been tapped also have 3/8" wiring holes for each cob. Most of them have been drilled and tapped along the side for mounting to an enclosure but I don't think it would be an impediment for any DIY job. A couple of them have broken tap bits in them but both bits are lodged in a hole on the side of the sink, broken off very close to flush.
 
NoFucks2Give

NoFucks2Give

Well-Known Member
Rahz said:
Anodizing only provides a small bump in performance
Click to expand...
Considering aluminum extrusion works well for natural convection as the extrusion minimum spacing and thick fins is similar to ideal natural convection fin spacing I have a tendency to associate extrusions with natural convection. With forced air it would be more effective to use a heat sink with more (and thinner) fins per inch.

A "small bump" is not importation if you do not sweat the small stuff. Personally I would consider 5-10% (forced convection) more than a "small bump".

What I have found is all the small stuff all adds up and in totality becomes big stuff. I tend be leery of someone that simply shrugs off a singular small parameter as they likely do the same in all their decision making where the end result is always less than optimum. In the end you end up putting on bandages to compensate. You can always add more air flow, decibels, and cost to the fans.

While on the subject of heat sinks, I have wondered about your ventilation. Why do you use such small and so few slits on your enclosure for ventilation?

But I guess you can over come all the small stuff with four big ass noisy fans in a small enclosure.
 
Rahz

Rahz

Well-Known Member
They're not that noisy if you buy the right ones. The 120mm fans I use move 75 CFM at 34dB. I built some custom rigs for someone using the 5.886 profile, 18" length with two CLU058-1825s at 50 watts each, 120mm fan in the middle. I didn't do any thermal testing because I knew they would be under 10C but there was nowhere on the sink you could touch that felt warmer than ambient. If used in conjunction with a tent/exhaust fan you wouldn't be able to hear the lamps unless you were in the tent. The 80mm fans I use are even quieter. Both the 120 and 80mm fans I use consume less than 2 watts each.

Anyway, back in the day most people were building fan cooled rigs using HSUSA sinks. Works well. The performance of any particular length/wattage will be more than a small bump in performance adding fans to these profiles used horizontally. Anodized only would require more material. End result would be heavier lamps with no real savings to justify it.

I've never released anything that didn't have exceptional cooling while at the same time I think I've done a good job keeping prices competitive and noise at a reasonable level. I hope this answers your question.
 
xX_BHMC_Xx

xX_BHMC_Xx

Well-Known Member
NoFucks2Give said:
Yesterday I attempted to calculate the watts dissipated (heat flow) per inch length of the 0.601" and 1.00" HSUSA heat sinks.

First off the problem with these heat sinks is they are not anodized. This is a huge problem for a heat sink. They do not work well (at all) if not anodized.

A new shinny unanodized heatsink has an emissivity coefficient of 0.04 (may as well be zero) where one that is anodized can have up to 0.95.
The perfect black body coefficient is 1.0.
This site seems to have fairly accurate ranges for emissivity coefficients: http://www.optotherm.com/emiss-table.htm

Also these natural convection formulas are for heat sinks with the fins pointing outwards and not upwards as the top and bottom areas between the fins needs to be open to gravity. Cooling depends on the density and buoyancy of the air.


The following were calculated using a web app done by Microelectronics Heat Transfer Laboratory, University of Waterloo, Ontario, Canada
In the attached ZIP file there is an html page filled out with the 0.601" parameters. Not a user friendly app. If there is an error you get no answer. No error code, nothing.
It's easier if you start with one filled out. But you can try with a blank form too: http://mhtlab.uwaterloo.ca/cgi-bin/rect_hs

It would be nice to find two heatsink calculators that match. Here's one with different results: http://www.coolingzone.com/flash.php?id=4


0.601", 1.2 Watts heat flow per inch View attachment 3982443


1.00", 1.98 Watts heat flow per inchView attachment 3982451
Click to expand...
According to the link you posted, these HSUSA heat sinks could have an emissivity coefficient of up to .3 considering they are in no way "polished" from the factory, and could even be scuffed a bit. Whereas it could be as low as .6 with an anodized heat sink. Not as dramatic if you don't take the stats out of context. If the cost of anodizing the heat sink is less than the cost of adding a bit more sink, then great. I find that in many applications it is more cost effective to use an extrusion than an anodized pin-fin, even when factoring in cost of fans and drivers. Sure, noise is a concern for many people, but my exhaust fan makes a hell of a lot more noise on its minimum setting than the fan on my light could ever hope to produce.
 
Profound Bastard

Profound Bastard

Active Member
NoFucks2Give said:
First off the problem with these heat sinks is they are not anodized. This is a huge problem for a heat sink. They do not work well (at all) if not anodized.
Click to expand...
That's just not true (at all).

Heatsinks work through radiation (which anodizing affects), convection and conduction. Last time I looked at this, in the sizes, temperature deltas, and airflow even passive LED heatsinks see inside of a grow room, the impact of radiation is minimized, probably less than 10% of the overall picture. Unanodized will and does work perfectly fine and there's no reason to scare people away from it. Radiation becomes a much larger factor in small heatsinks relying solely on natural convection, like you see in electronics.

Also, heatsinks are really fucking complicated.
 
Last edited:
NoFucks2Give

NoFucks2Give

Well-Known Member
Sorry I had to leave before I finished my post and did not fact check. This site has too short an edit period to fix things.
I will retract my statement "This is a huge problem for a heat sink. They do not work well (at all) if not anodized." (Forgot I said that)

I will agree anodized or not is not a BIG issue for most forced convection heat transfer.

For natural convection it is a bigger but also more of a conditional issue.

I sweat the small stuff. I look at every little detail as I have found it pays off big in the end. For me, I will continue to go by my rule to only use anodized aluminum in the cases were I would use aluminum. I'm a copper guy.

xX_BHMC_Xx said:
the cost of anodizing the heat sink is less than the cost of adding
Click to expand...
Cost of anodizing should not be an issue for @Rahz He has an enclosure, it should be aluminum, aluminum sheet metal needs to be anodized, so adding the heatsinks to the sheet metal anodizing should not increase cost much if at all. I do however still have an issue with his enclosure ventilation.

In my fact checking I found this summation on the topic of anodization: https://www.aavid.com/product-group/extrusions-na/anodize
I said 5-10%, they say 4-8% in the case of forced convection.

Profound Bastard said:
really fucking complicated.
Click to expand...
We are on the same page with that. It is much more empirical than analytical.
 
Rahz

Rahz

Well-Known Member
NoFucks2Give said:
I do however still have an issue with his enclosure ventilation.
Click to expand...
Any yet all you have to offer is an opinion you can't substantiate. Not the first time this has happened. The cooling solutions on my lamps work just fine. And let's be honest about this thread. We weren't talking about my lamps, we were talking about HSUSA heat sinks. Your issue isn't with my lamps, it's with anyone who contradicts anything you have to say. These horizontal sinks with the wattage we use on them need fans. I wasn't trying to be an ass. I was just making a valid point and you respond by disparaging my products. Good work.
 
Last edited:
xX_BHMC_Xx

xX_BHMC_Xx

Well-Known Member
Since it seems we're on the topic, what is the issue with the ventilation on those lights? I see no problem there. Seems to be plenty of room for air to escape. Also many people own those lights, I haven't heard about any overheating. And anodizing parts that don't really need to be anodized adds cost to a product. If Tasty lights were going for 3 times what they are, I would expect anodized heat sinks.
 
NoFucks2Give

NoFucks2Give

Well-Known Member
Rahz said:
Any yet all you have to offer is an opinion you can't substantiate. Not the first time this has happened.
Click to expand...
I mentioned that ("I do however still have an issue with his enclosure ventilation") because I had previously asked and you failed to answer.


I was just wondering why the slits were so small. You must have had some reason to make them that size.
Forced convection is about air flow. It is possible there was some logic behind the reason. I failed to see it. I was just asking why.
I though because you use an AC plug and likely lethal voltages, maybe your safety certification dictated. But I did not find any info regarding safety certification's on your website. UL usually specifies the size of openings based on the voltages used and the type of certification.

Not that I previously offered any opinion.

Rahz said:
We weren't talking about my lamps, we were talking about HSUSA heat sinks.
Click to expand...
It was you that brought up your lamp.

xX_BHMC_Xx said:
what is the issue with the ventilation on those lights? I see no problem there.
Click to expand...
My remark was not about whether the enclosure was adequately ventilated because you can increase the CFM rating on the fans used. It just seemed to me that improved ventilation would reduce noise and cost of the fan by being able to reduce the fan's CFM. Just another one of those "small bumps" that add up that can be cured with a big ass fan. Just trying to help.

Size of the slits is a major factor in forced convection within an enclosure:

airFlowFormula.jpg
 
Rahz

Rahz

Well-Known Member
I did not bring up "my lamps" I brought up using the 5.886 profile from HSUSA which is entirely relevant to this thread. It has nothing to do with the lamp in question. You said "while we're on the subject..." but I already know from previous conversations that your intentions are shifty and have no desire to answer your questions.

In that very same post your offered opinions were that I use big ass noisy fans, and have small and few vents, and you insinuated that I'm likely to shrug off all parameters in my decision making process. You did this not out of any genuine concern or interest but because I offered information that contradicted your post regarding anodizing and you felt threatened. It's the exact opposite of your screen name. If you were really concerned about the size of my vents (small and few when applied to a value of more that three are both subjective opinions) and sound levels of my fans (you had no idea what the dB rating on them is) you would have just asked but instead you offered disparaging opinions. And now you're trying to claim you didn't offer opinions and that you didn't bring up my lamp. That's called lying. And when it comes to proving your opinionated assertions (for instance that Alesh's spreadsheet has a bad formula in it, or that my lamps are loud and don't have enough ventilation) you got nothing to give.

Again, there is plenty of ventilation and the fans are reasonable quiet. I could go into detail but you're a troll and I know you couldn't care less about my lamps.
 
NoFucks2Give

NoFucks2Give

Well-Known Member
What I was originally trying to get to was the purpose of the tables.

Temperature Correction

Thermal resistance of a heatsink is typically specified at 75° C.
Natural convection is temperature sensitive. As the temperature delta between the heatsink and ambient decreases the effectiveness also decreases.
Example: if the thermal resistance of a heatsink is 10°C/W at 75° it will be 11.6°C/W at 50°. (from table, 50° factor = 1.106) 10 x 1.106 = 11.6

Length Correction
The C/W/3 approximated thermal resistance of an extruded heatsink is the specified at a length of 3 inches.
When the length increases beyond 3 inch he effectiveness of the heatsink increases as well.
Increasing the length of a heatsink has diminishing returns.
You simply multiply the C/W/3 rated thermal resistance of the heatsink by the length factor.

Power Factor
This table I assume is mislabeled. There is no such thing as Power Factor, that I know of.

Where Length Correction is for natural convection I assume the "Power Factor" is similar to the Length Correction but for forced convection.
I could not figure out what value it is normalized to. I calculated heat flow values for various conditions but nothing panned out.

These "factors" are very similar in magnitude when compared to the natural convection Length Corrections.


Best I can tell this spreadsheet appears to use surface area to determine the heatsink's effectiveness. That would not work.
Another problem is the thermal resistance stated by the manufacturer is highly dependent on the heat source used to calculate its value.
The thermal resistance would only be valid for the heat source used by the manufacturer. A heat source with different dimensions and wattage will create different temperatures and these calculations are very temperature sensitive.
 
NoFucks2Give

NoFucks2Give

Well-Known Member
Rahz said:
I did not bring up "my lamps"
Click to expand...
Seemed like you brought up your lamp in regards to the anodizing. In my opinion.

Rahz said:
Anodizing only provides a small bump in performance when cooled via forced convection. I was achieving 10C over ambient using the 5.886 profile with fans, about 4.15 watts of heat per inch.
Click to expand...


Rahz said:
I offered information that contradicted your post regarding anodizing and you felt threatened.
Click to expand...
Oh. I didn't think you contradicted what I said about anodizing. I do not understand WTF 4.15 Watts per inch meant, kind of a strange way to rate a heatsink. I thought you bandaged your problems just fine with your big ass noisy fans. 10°C over ambient is okay. I did not feel threatened. I apologized for not fact checking and retracted and clarified the part about the magnitude of anodizing. But the part about the "small bump" still stands. It does not matter what I think, I cannot buy a product without a UL label.

That's all this liar and troll has to say to you.
 
You must log in or register to reply here.
Top