Please add me to the interest list.
Yeah, metal around the head, well turbo time should grow pretty much in proportion to weight there, and fins, even better. It takes twice as much heat to bring twice as much metal to the same temperature. Anyway, it's not much and it's fine, I think both issues sound like they're getting pretty small. Half of me wants to buy two, but it's not "the better half".
Yeah will update list later
Flintrock, if half of you want 2, I must put you on the list for 4 right 
My basic rules (I think most modders/designers would agree with) with some cost consideration for passive cooling in flashlights:
- limit the # of thermal junctions: metal to metal, threaded connections, use one piece construction, etc.
- metal to metal junctions should be as close and large as possible (well clamped, smoothed surface flat contact, etc.)
- where there are thermal junctions, increase their efficiency (solder, thermal grease/epoxy properly applied, etc.)
- copper is best closest to the heat source, aluminum has it's advantages in shedding heat out (might not all agree here), maybe aluminum is just more practical for this than copper, for example
- the more surface area exposed to the air in the outside shell, the better (more fins, more size, more...)
- location, location, location - get the material mass and cooling fins closer to the source
- active cooling would be great, but practical implementations abound
Take these rules and apply practical limitations and trade-offs of cost, portability, functionality, styling, etc.
Mostly that looks right.
You don't touch on is mass much though. Heat transport is one thing, but heat capacity is another. Mass doesn't impact sustainable output much except as it affects other the other things (thickness of thermal path etc), but it surely impacts turbo time. Aluminum can hold almost 1 Ws/(g*K) so at for 50K rise, 50g more mass will hold about 50s of heat at 50W. Copper by the way has a much lower heat capacity by mass, for better or worse though it makes up for it by weighing more.
True, just mention the location of the mass, not the amt of mass - need to update the list 
I like the premise of the classic CPU heat sinks - copper inside at the chip, aluminum to the air.
Others know the details better, but I think/thought copper is great for pulling heat, aluminum can conduct it quicker. Measurements and specs from SinkPAD and tests done here on BLF show copper MCPCB's perform better, but not by any big margin, and pflexpro here has some really interesting test results for P60 configurations that has some general use applications.
Updated list:
- limit the # of thermal junctions: metal to metal, threaded connections, use one piece construction, etc.
- metal to metal junctions should be as close and large as possible (well clamped, smoothed surface flat contact, etc.)
- where there are thermal junctions, increase their efficiency (solder, thermal grease/epoxy properly applied, etc.)
- copper is best closest to the heat source, aluminum has it's advantages in shedding heat out (might not all agree here), maybe aluminum is just more practical for this than copper, for example
- the more surface area exposed to the air in the outside shell, the better (more fins, wider, spread it out)
- bigger is better - I’m all about that mass, bout that mass, no treble
- location, location, location - get the mass and cooling fins closer to the source
- active cooling would be great, but practical implementations abound
Let’s do some science!
The specific heat of aluminum is ~0.900J/(g K) and the density is 2.70g/cm^3. The specific heat of copper is ~0.385J/(g K) and the density is 8.96g/cm^3.
Let’s say a S2 triple spacer is 3.8cm^3 (because kiriba-ru lists his cu spacers as 34g). An Al spacer will weigh 10.26g and a Cu spacer will weigh 34g.
An S2 triple is say, 35W heat energy needing dissipation (35W = 35 J/s). The allowable Temp increase over ambient we define as 50*C.
Q=c*m*(T2-T1)
For Aluminum: (35J/s)(X)=(0.900J/gC)(10.26g)(50C) —> X=13.2s
For Copper: (35J/s)(X)=(0.385J/gC)(34)(50C) —> X=18.7s
So for a fixed _volume _of metal (which is basically the constraint we are facing), the Aluminum will heat up to our predefined “limit” temp in 13.2seconds, compared to 18.7s for copper. The copper will take 42% longer to heat up to the same temperature.
Copper is of course be three times heavier, so it is up to the user to subjectively decide if copper is worth the weight penalty. But objectively, copper is the better heatsink material without argument (I doubt anyone is arguing this).
Note that this calculation is purely regarding the heatsinking ability of the metal, which is not the same as it’s ability to dissipate that heat to the environment. As Flintrock points out, the heatsink calculation primarily (but not only) affects the light’s initial ability to maintain it’s output in Turbo before output starts dropping due to heat buildup.
The ability of the metal to then shed that heat to its surroundings (either in continuous use, or after the light has been turned off or reduced in power, it doesn’t matter) is a different calculation, where the thermal conductivity of the metal is the critical variable – and copper has a higher (better, in this case) thermal conductivity than aluminum. So for identical conditions/environments/geometries (i.e. apples-to-apples), copper will shed heat to its environment quicker as well, and thus perform better in this regard as well.
@ sac02………
I honestly love to see these calculations where science backs up or disproves theory/opinion/speculation.
There was a time long, long ago where I understood a “tiny” bit of the calculations… but life took another road. 
Thanks for posting that sac02…. :+1:
.
To add another factor involved in regard to what happens in a flashlight:
Copper is about the best heat conductor in existance, while aluminium, though twice as bad, is still very very good at it. The amount of heat that is conducted is also a function of cross section area the heat has to go through which is increasingly smaller while closing in on the heat source. That is why very close to the led (i.e. the solder pad) using copper has a large effect, while further away it becomes ever so irrelevant to use copper or aluminium.
There were many pages of interesting discussions on copper/ALU/stainless steel in the Kronos X5/6 topic.
Made me Google and anodised ALU is better in shedding heat then bare. I think even color plays a part but that is info from much longer ago when I liked doing crazy cooling in computers.
But we are all in agreement I think
The heat “holding power” of aluminum, and mass, is easily demonstrated. And most folks do it frequently.
Put something in a hot oven to cook on a piece of aluminum foil. Take it out and one can immediately lift it by the loose ends of the foil. It had temperature but not enough heat stored to burn skin. Note that it’s foil and not an aluminum pan. Add mass in the form of a pan and all of a sudden it contains plenty of stored heat to burn.
Edit: I would never try that with copper.
Would it be more accurate to refer to the increasing temperature gradient as you near the LED, instead of cross sectional area? I think I understand what you are trying to say with the cross-section analogy, but the driving force is the temperature gradient (which as you say is lower, farther away from the LED), right?
OK, digging deeper into my memory (I don’t still practice thermodynamics regularly, this is stretching the bounds of my memory) and refreshing myself with some Wikipedia, Fourier’s law of heat conduction is proportional to both temperature gradient and cross sectional area. But I’m thinking that’s applicable to say, the LED heat pad vs the MCPCB base (very small area at the LED pad limits the heat flux at that point, larger area at the MCPCB allows for more conductance across that plane). Within a homogeneous material, I think the mechanism is thermal diffusivity, which is a function of thermal conductivity, density, and specific heat.
Actually, as I refresh more, thermal diffusivity appears to be the characteristic that I was kind of getting at with the above calculation using density. (nope, that would be volumetric heat capacity. ) Only precious metals have better thermal diffusivity than copper, Copper is moderately better than aluminum, and aluminum is significantly better than any other common material.
So one thing that I think could become an issue, I see so many posts saying that a specific light gets too hot to hold very quickly and this is seen as a negative aspect of the light. From a thermodynamics point of view, this means that the design to transfer heat away from the LED’s is very efficient and the diodes can sustain turbo much longer, but larger mass will help to delay the raising temp of the body. If the designers of this light do too good of a job, I bet there will be many complaints that it gets too hot lol.
Regarding color to thermodynamics, black anodize is about the best you can get on the aluminum (closer to a black body - ideal radiator), but that only really affects the radiation heat transfer which is minimal compared the the convection (air) and conduction (hand).
BUT, this is really fun engineering/physics discussion that will never matter due the variances in manufacturing tolerances of the LED/battery/driver components, ambient air temp, user hand size, phase of the moon, etc. so please don’t let the shelf thickness issue become an issue!
Just my $0.02…
heheh from watercooled PC to bloodcooled Q8 
Perhaps this was already discussed in this giant thread ( ), but who is going to distribute the lights? I see that Neal was on board, but since he left I was wondering if BG is still on board?
This is a good fun discussion. :+1:
My question is this, good common easy to understand by anyone examples were given for convection & conduction.
What is an example of radiation?
Heat transfer……
- Radiation =
- Convection = Air
- Conduction = Hand
banggood is on board for the sales
Cool, thanks
I am beginning to think we have a lot of the crazy drivers from Austin TX here on the forum, impatient and willing to crash people just to get moving again.
FWIW, the light I recently built from scratch cost me some $175 in materials and a little over 20 hours in labor, yes it makes almost double the lumens compared to the Q8, but the point really is if you try and do it yourself it’s going to get really expensive.
<—- Bowing to the Master’s and Queen codewriter’s here, standing by for the final product with zip for complaints about how it’s all going….
Also, FWIW, I’ve carried MILLIONs of pounds of construction materials, don’t really see where softball gives anyone much added strength although I guess a desk sitter can gain something from it. I worked with Ernest Dye when I was a young man. This dude could, at 67 years old, take my Dad’s belt in the middle with one hand and put Dad (around 148 pounds back then) to the ceiling… entirely one handed! 60 years old is no excuse for being unable to do much of anything. Just saying…
DB - huh?
Teacher - radiation heat transfer occurs between two objects without the medium in between being involved, the heat transfer is between the two objects. For example you can feel heat from a campfire several feet away even if the air is cold. The sun can warm a surface even though there is vacuum of space. You could put a hot flashlight on Turbo and a cold light into a vacuum chamber with no air, and the hot light would heat the cold light even without any air to transfer the heat between the two (that would be convection).