3DVelocity would like to thank Swiftech and especially Gabriel Rouchon (Swiftech Founder, President, CEO) for their help and courtesy in providing this cooler for review.

A Closer Look :

Let's kick off with a look at the fan. The unit Swiftech decided to use for its stock configuration is the Pabst NHH612. This really is an excellent fan that offers 33CFM of air shifting muscle at a rotation speed of 6850RPM. Surprisingly however, it draws only 3.0watts and although the noise levels are recorded at 43DB, it's certainly a lot quieter than the infamous Delta. The lack of RPM monitoring is a problem in my book, but other than this the Pabst is an excellent fan that boasts incredible performance and has a superb reliability record.

 

Swiftech have opted to mount the fan using four long (2.5") Phillips headed bolts that thread into the base. In addition to the fact that this essentially adds four extra cooling pins to the array that's already there, it also eliminates the need for unnecessary shrouds or mounting brackets on top of the sink itself. The holes in the base are threaded all the way through allowing a little flexibility in the width of fans that you can fit on top should you decide to swap for another.

You'll also notice that the fan sits on rubber spacers. The main reason for this is to stop fan vibration transmitting noise through the heatsink and into your case. I can't see this being particularly effective as the four bolts would transmit the vibration anyway, but there are no complaints from this reviewer regards noise levels so I guess every little helps. It is also claimed that the rubber spacers raise cooling performance very slightly, and the theory behind this is linked to what's called the "dead zone" or "void"

The dead zone is the area directly below the motor on a fan. Because the motor on high performance fans is often quite bulky it actually obstruct airflow down into the heat sink, producing a pattern of air that initially emerges from the fan as a hollow cylinder. The theory is that by raising the fan, the air has time to spread a little before it initially reaches the cooling surfaces thus reducing the size of the dead zone. The diagram below is a very simplified representation of the dead zone theory as it doesn't take into account thinks like turbulence and interference, but it does show an approximation of what occurs.

Mounting a Swiftech is simplicity itself. On each side of the cooler are mounting clips which provide pressure by means of a spring. To fit the heat sink, first you drive in the screw. This lowers the clip by compressing the spring. With both sides screwed down the clips slip easily onto the socket lugs. To apply the required pressure you simple need to slacken the screws until the heads are clear of the top of the clips and you automatically have the correct downforce. It is important that each clip be slackened alternatively a little at a time or the pressure will be applied to one side first which may damage the die. Also, because of the spring pressure the clip tends to twist a little while you're slackening the screws so you need to keep checking they're still located properly on the lugs as you go. In accordance with AMD and Intel Heat Sink to CPU pressure specifications, the two springs are of unequal length so that the die, which does not sit centrally against the base, receives equal forces from both sides and thus sits perfectly flush.

 

In addition to being one of the most die friendly mounting methods around, Swiftech's unique mounting clips offer other advantages. As you can see from the picture below, the mounting system means Swiftech have been able to eliminate the usual fin-free zone across the centre that's needed to accommodate regular mounting clips. This means that there are now cooling pins where they're most effective, right above the core. For the record, the cooling fins are press fitted into the base rather than soldered.

But were Swiftech happy to leave things like this? Not on your life! They decided they wanted to effectively double the surface area of each of the 188 cooling pins and found a very clever way of achieving it. By tapping a helicoid thread onto each pin, they have literally doubled the effective surface area which now checks in at an impressive 231 square inches. This helicoid design also creates an amplified turbulence profile that can increase the effective velocity of airflow in places and keeps the air in contact with the cooling surfaces for fractionally longer.

I hinted earlier that the base was very flat, and I wasn't kidding. Swiftech claim a base flatness of better than .001", so I took them up on it and had it checked by a good friend who runs an engineering company. Not only did he confirm that the base on our review sample was within this tolerance he also remarked about the high quality of the finish. Okay, it's not quite mirror smooth, but it's certainly not a candidate for lapping, and because of the way thermal greases like Arctic Silver are constructed, a mirror finish can actually be less effective as it offers no key for the suspended particles.

The procedure for getting such a high quality base finish was explained to me by Swiftech Founder Gabriel Rouchon. Apparently the first step in the process is to have the bases professionally machine lapped on a UK built machine known as a "Lapmaster". Rather than stopping here Swiftech then continue the process manually on a granite polishing table using a special polishing paper. The final MSF (Micro Surface Finish) rating for the base is 8, while a true mirror finish would be classified a 1. For reference, even the better HSFs available average an MSF rating of 16 with the worst examples hitting the 30-40 mark.

 

Page 3 - Results and conclusion>>>