Design and Functionality
The first impression from Noctua NH-D14 is definitely made by the size and weight of this product. The cooler measures 160x140x158 mm and weighs 1240 g with two fans, 1070 g with one fan and 900 without any fans. At the same time, the cooler doesn’t look bulky or clumsy. On the contrary, the solution made of two tower heatsinks has a pretty strict and finished look:
Noctua NH-D14 is built on six copper nickel-plated heatpipes 6 mm in diameter. The heatpipes go through copper nickel-plated base and pierce two heatsink arrays, each consisting of 42 aluminum plates with 0.45-0.50 mm thickness.
The gaps between the plates are 2.5 mm and the space between the heatsink arrays measures 28 mm. Therefore, you won’t be able to install thick fans with high static pressure between the heatsink arrays.
Note that the heatsink arrays aren’t symmetrical in respect to the cooler base: the second array along the airflow path is slightly shifted forward. It must have been done in order to make the heatsink with installed fans smaller, because the air intake fan should be installed onto that part of the heatsink that is closer to the vertical base axis. Although in this case it is not all that simple, we are going to get back to this matter later when we discuss the benchmark results.
Heatsink arrays are absolutely identical in shape as well as plates size. Each plate measures 140x51 mm. The heatpipes within each heatsink array are placed into a single line without any special “optimizations”:
The plates are soldered to the heatpipes. The calculated effective heatsink surface of Noctua NH-D14 is about 12,020 cm2 and it is the best number among air super-coolers (Cooler Master V10 and Scythe Mugen 2 have about 10,660 cm2 and Thermaltake IFX-14 – 10,320 cm2.
The sides of the heatsink plates are notched to lower the airflow resistance:
As you may remember, the heatsinks of Noctua NH-U9, NH-U12 and their modifications also boasted similar plates optimization, however, it has become significantly more distinct and most likely more efficient in Noctua NH-D14. At least, I suggested that it would be a good idea for Thermalright engineers to apply the same notching to the sides of their cooler heatsink plates about a year ago, but unfortunately, they didn’t seem to pay any attention to that suggestion.
The heatpipes are also soldered to the cooler base, but it is done so neatly that you can only notice traces of soldering once you take a real close look at the cooler:
Noctua have never pursued mirror-shining base surfaces. And today is also no exception, as we see no polishing of any kind on the NH-D14 base:
I don’t think we should consider it a drawback. However, the issues with the base surface evenness definitely cannot be overlooked:
The contact surface of the base obviously has a small bump shifted towards one of the sides. Hopefully, it is typical only of our specific sample and not the entire lineup, unlike the solutions from one of the popular cooler makers. If all Noctua NH-D14 cooler will suffer from this issue, the users will have to take matters into their own hands and even out and polish the base on their own.
Despite the not so idea surface of our Noctua NH-D14 cooler sample, the thermal compound imprint off the processor heat-spreader turned out quite OK:
Our regular readers should be well familiar with the first fan, however the 140 mm one is a new kid on the block. It is new not only because it is larger and can be installed into the same retention holes as the 120 mm model, but also because it uses a rotor of modified design with metal reinforcements that should increase the rotor and fan MTBF. The remarkable thing is that there is a metal cap on the rotor instead of the common paper or film sticker:
We are going to take a closer look at this new fan alongside with a bunch of others in the upcoming fan roundup. At this point I would only like to add that the maximum rotation speed of the Noctua NH-D14 fans are 1300 RPM for the 120 mm NF-P12 model and 1200 RPM for the 140 mm NF-P14 model. You can use the enclosed rotation speed adapters to lower the speed of both fans to 900 RPM and the bundled Y-splitter to power them from the same mainboard fan connector. All other fan specs will be listed in the table below.
I have to point out that the fans do not touch the heatsink but sit on eight silicon bricks inserted into heatsink grooves. Thanks to these bricks the cooler generates less noise and the fans are raised 3-4 mm above the heatsink arrays. Together with notched sides of the heatsink plates this reduces the dependence of the cooler efficiency on the fan rotation speed. There is one thing, however, that could be improved: if the silicon bricks were only one third of their actual size, they wouldn’t cover so much of the heatsink plates. Any user can easily do it with a pair of scissors.