Thermal properties of metals at a glance

Like a true flashaholic i’ve been interested in the different thermal properties of the metals used in lights these days. I understand most people have come to the conclusion copper is best followed by aluminium and brass then just about everything else. But I haven’t found an easy to read at a glance comparison between the commonly used metals. When comparing to copper even brass looks poor. But which grade of brass or copper. Some thermal characteristics can change quite a bit when alloyed with other metals.

Half the problem is I don’t actually know which grades of metal are used. I suspect 6061 aluminium, C101 copper, 6al-4v titanium, 316 stainless and brass …???…

So after some googling I put together this graph to capture at a glance the difference between the thermal properties of some common metals. If you see an error then you’ll know it’s because I just googled it. The only surprise to me is ti and ss are as bad as each other, for some reason I thought ti was marginally better than ss. Copper also looks to be a solid performer no matter the grade.

Aluminium

1000 – 230 W/m-K

5052 - 138 W/m*K

6061 - 151-201 W/m-K

6063 - 201-218 W/m*K

7075 - 130 W/m-K

Copper

C101 (99.9%) - 391 W/m-K

C106 (standard commercial grade) - 391 W/m-K

C110 (99.99%) - 390 W/m-K

Titanium

6al-4v (grade 5) – 7 W/m-K

all grades are poor – 7-22 W/m-K

Brass

C360 – 115 W/m-K

73/30 – 120 W/m-K

Stainless steel

316 – 16 W/m-K

304 - 16 W/m-K
.

.

I remember reading a thread a few years ago of somebody testing the thermal aspects of al vs cu. From memory the al took slightly longer to (heat) soak, was only a couple degrees hotter but maintained that small difference until the temperature of the heat source was changed. I don’t actually remember reading it but I suspect the al cooled faster than the copper. My guess is it did because the lighter density of al does not ‘hold’ as much heat as copper. When a hand warmer becomes a hand burner I want it to cool as fast as possible when it’s turned off.

Of course, the thermal properties of a light is only one facet of the many aspects of a light. I’m not a big fan of copper host lights but I am a big fan of dtp copper mcpcb’s. A copper host is not the best choice when using a brass pill or regular mcpcb.

Anyway I hope someone else finds this interesting. Copper reigns king, aluminium is good, everything else…. meh. For reference here’s another related table I found it shows even more metals - Thermal Properties of Metals, Conductivity, Thermal Expansion, Specific Heat

Good write up! :+1:
These be the more common materials used, the more exotic alloys, like CuSil or just Silver, Artificial Diamond Wafer and Graphene, the wm*K starts shooting up.
One of these days, someone is gonna figure out how to use heatpipes within some of the larger light heads to further dissipate the heat these ever more powerful LEDs and drivers are cranking out. (Not like the Imalet MS18, they just used a bog standard cpu heatsink in there somehow)

The thermal conductivity is similar to electrical conductivity in that silver is the best conductor, followed by copper, then gold.

But the property of Heat Capacity or Specific Heat relates the amount of energy that can be absorbed to the temperature change that will be experienced. Magnesium has the highest HC, but aluminum 6061 is more commonly available and makes the best heatsink or housing. It can absorb a huge amount of energy before “heating” up to the touch, then it cools quickest when the light is turned off.

Copper transfers heat better than aluminum, just like it has lower resistance in wire. But once the energy has been absorbed and the temperature goes up, then it will stay there a long time (slower to cool down) due to the lower heat capacity.

Darn. I have been looking at a Brass light (because I’m not fond of copper patina) because I though it would be closer to copper. Being worse than Aluminum was not what I expected.

Thank you for the list of metals. Didn’t know that titanium and stainless steel are that awfully bad thermal performers. Only 5-10% of aluminium! :disappointed:

We’re thinking small, boys. I want Synthetic Diamond:

Also, there is a Table here.

More Exotic:

More exotic wont pass ROHS, as its a toxic compound

Moissanite diamonds can be produced cheap with similar properties to pure diamonds

Sounds like a Cu head would double best as a hand warmer for winter walks etc…

Yes, it relates to the total amount of energy transfer but not the speed at which it can happen. The speed it transfers is measured by thermal resistance.

If you ever come across a diamond, silver or gold flashlight please show us :wink:

Once the light is turned off it will stay warmer for longer :sunglasses:

Please note that thermal conductivity is one of 2 big properties. The other is specific heat capacity i.e. how fast does it heat up.
You already noted that:

There are 2 kinds of specific heat capacity; the most common is gravimetric i.e. “how much heat does it take to up the temperature of 1 gram of substance by 1 K”. For us the most interesting is volumetric i.e. “how much heat does it take to up the temperature of 1 m³ of substance by 1 K” though both matter. Here heavier materials (like copper) have an edge.

Heat capacity, like conductivity, depends a lot on alloy. From my old notes:

Regular alu 6061 has thermal conductivity of 150 K/W and heat capacity of 2.42 J/mm³K. 6063 has 200 K/W and nearly the same 2.43 J/mm³K. Copper has 400 K/W with 3.44 J/mm³K. 7068 which I think is the overall best flashlight alu for being much tougher than 7075 with great thermals – has 190 K/W with 3 J/mm³K.