The Noctua has significantly better cooling performance; i did test it against my Hyper N212+ for AlienBabelTech - and out came the CM cooler
Machined bare copper is poor for conducting heat; you have to put on the thermal compound with a spatula - like some women do with pancaking makeup over their blemishes and deep holes (OK, i exaggerate)
:biggrin:
If i am not mistaken, Nickel plating thickness is measured in microns and it keeps the copper from oxidizing - the Noctua base is like a mirror's perfect surface in comparison to the rough surface of the N212+
^_^
Well, we've been all over this over the past four years, and you're free to disagree anyway. I ran experiments, got my CoreTemp posting to a log-file every few seconds for runs of several hours.
Several metals/materials/substances used in heatsink construction are chosen for their thermal conductivity and thermal resistance. I think you'll find tables published in archives for OC'ing sites [
www.overclockers.com is one . . ] with this data. Aluminum -- among heatsink choices, ranks at the bottom of those used. Then follows copper. Then silver. I forgot where gold fits in, but it's worse than silver as I recall.
As you noted, they use nickel-plating to inhibit corrosion. There is another table called the "Galvanic" table, and elements/materials closest to each other are often chosen to plate another for that reason -- Nickel and Copper are together on that table. But Nickel has inferior thermal conductivity and higher thermal resistance.
My own experiments resulted in bar-chart frequency distributions for "plated" versus "lapped-to-bare-copper." They confirm what several of us here might agree: you can reduce your high-temperatures by grinding off the nickel from a copper IHS or heatsink base, and the temperature reduction is additive for two surfaces.
As for corrosion, I spoke to ThermalRight tech-reps and it became clear that their biggest concerns were corrosion or compromise to the welds or solder-joints where the heatpipes meet the base. If the area exposed will be covered with TIM and sealed under pressure, it's not going to result in the situation you might find sticking polished copper on a pole at the beach to expose it to salt in the air.
Earlier in recent weeks here at the forums, we concluded that the plating process is done for the final assembly as you'd imagine it, and that lapping the base would add to manufacturing costs, so they don't do it. And of course, lapping could be done by the uninitiated grasping the assembly by the fins and pipes, which might result in damage to the joints, so -- of course -- they would take the opportunity to include that in things voiding a warranty.
The only problem I see with the CM cooler is the gaps and grooves. I also had a Xigmatek "direct-touch" cooler, tried it, and found that it didn't perform as well as a lapped ThermalRight, Noctua or Prolimatech, although in theory, it should have an advantage for removing the base-plate from the equation.
But in my judgments here, I can see a lot of these coolers as "more than acceptable." You'd just want to pick the one with the lab-measured minimum thermal resistance in published results using identical test-beds. You can trade-off expense against peformance given the range of prices, but if you do so, it should be a deliberate choice to compromise a few degrees on the load temperatures.
On the other hand, if the market assigns prices based on performance as well as quality and construction factors, you have a choice of paying maybe an extra $10 or $20 to capture a few degrees Celsius over other options. [Of course, sometimes [more often than some may think!] markets don't work efficiently, but efficient pricing is more likely in this near-perfectly-competitive situation of several . . . many different manufacturers. Any anomalies will be less evident than a fairly good ranking according to "get what you pay for."
Facebook
Twitter