robincnn
Well-Known Member
Summary: Under active cooling Anodization does not help and may degrade performance.
I have not seen any anodized Active cooling CPU heatsinks anyway. They would have made one by now if anodization was reducing temperatures under active cooling.
I came across this article and it said anodizing helps reduce temp by only 6% in case of active cooling.
www.rollitup.org/f/led-and-other-lighting.124/create-thread
Given the thermal conductivity of Anodized layer and dye I thought it could actually degrade the temperature of heatsink.
https://www.rollitup.org/t/does-anodized-layer-impact-heat-transfer.869784/
I tested two identical heatsinks. One regular, other anodized. It may look a crappy job with anodizing but its only bad dye. The heatsink is no conductive and has emmisivity as high as black tape so I am pretty sure it is a decent anodization.
Regular heatsink - LUXEON XF-3535L strip 3000k, 140lm/w
Anodized heatsink - LUXEON XF-3535L strip 4000k, 147lm/w, Anodised layer was removed for mounting area with 150 grit sandpaper and then 600 grit.
Due to different color temp and the way led was mounted there is some bias.
Ambient temp = 22 degrees
Temperature checked with IR gun with emissivity set to 0.95
Passive cooling (Radiation cooling dominant)
Active cooling (Convective cooling dominant)
I assume this is because passive cooling is proportional to Temperature to the power of 4, so it is effective at higher temperatures.
With active cooling i don't know the formula but it seems the rate at which convection can cool the heatsink is higher than the rate at which anodized layer/dye can transfer heat from inside heatsink to boundary of heatsink.
I will post Front side pics of heatsinks, exact tempratures and exact power of each strip. I think the power 4000k strip on anodized heatsink was 13% less, 5% more efficient, but still it was more hot than 3000k strip.
Disclaimer
Results be vary at different fan speed and there is some bais in data as i mentioned earlier. Still it seems pretty obvious that we don't need to anodize heatsinks if we are using active cooling.
I have not seen any anodized Active cooling CPU heatsinks anyway. They would have made one by now if anodization was reducing temperatures under active cooling.
I came across this article and it said anodizing helps reduce temp by only 6% in case of active cooling.
www.rollitup.org/f/led-and-other-lighting.124/create-thread
Given the thermal conductivity of Anodized layer and dye I thought it could actually degrade the temperature of heatsink.
https://www.rollitup.org/t/does-anodized-layer-impact-heat-transfer.869784/
I tested two identical heatsinks. One regular, other anodized. It may look a crappy job with anodizing but its only bad dye. The heatsink is no conductive and has emmisivity as high as black tape so I am pretty sure it is a decent anodization.
Regular heatsink - LUXEON XF-3535L strip 3000k, 140lm/w
Anodized heatsink - LUXEON XF-3535L strip 4000k, 147lm/w, Anodised layer was removed for mounting area with 150 grit sandpaper and then 600 grit.
Due to different color temp and the way led was mounted there is some bias.
Ambient temp = 22 degrees
Temperature checked with IR gun with emissivity set to 0.95
Passive cooling (Radiation cooling dominant)
- Regular heatsink - 50 degrees on black tape. 30 degree all over heatsink other than black tape, which is wrong reading due to emissivity set as 0.95 and aluminum having emmisivity of around 0.2
- Anodized - 40 degrees all over heatsink and same temp over black tape.
Active cooling (Convective cooling dominant)
- Regular heatsink - 26 degrees on black tape.
- Anodized - 27.5 degrees all over heatsink and same temp over black tape.
I assume this is because passive cooling is proportional to Temperature to the power of 4, so it is effective at higher temperatures.
With active cooling i don't know the formula but it seems the rate at which convection can cool the heatsink is higher than the rate at which anodized layer/dye can transfer heat from inside heatsink to boundary of heatsink.
I will post Front side pics of heatsinks, exact tempratures and exact power of each strip. I think the power 4000k strip on anodized heatsink was 13% less, 5% more efficient, but still it was more hot than 3000k strip.
Disclaimer
Results be vary at different fan speed and there is some bais in data as i mentioned earlier. Still it seems pretty obvious that we don't need to anodize heatsinks if we are using active cooling.
