Most recently, I introduced you to the basic information about what a water block should be for a Ryzen processor with a Zen 2 microarchitecture. For those who have not read, I will remind you that for processors with a Zen 2 and Zen 3 microarchitecture, any water block will not work. And if it is physically compatible with the AM4 socket, this does not mean that this sample will have optimal performance. The main reason is the shifted CCD relative to the center, as well as the increased heat flux relative to the crystal area. It seemed that the limits of performance had already been reached and all bicycles had been invented, but the German company TechN appeared on the horizon, which stated that their product was not only more productive than competitors, but also cheaper. Usually I am skeptical about such statements and like to refute it. In today’s material, I approached the issue of testing with the utmost responsibility: five water blocks, MO-RA3 420 radiator and, of course, Aquacomputer Aquaero 6 with a large number of premium sensors. It will be interesting, let’s start!
TechN is new to the water block market, however, it has already managed to develop water blocks not only for Intel and AMD processors, but also for video cards (in particular, for the Nvidia GeForce RTX 2080 Ti). It is usually extremely difficult for new players to compete with companies that have been on the market for 10 years. You might think that there is no chance, but not in this case.
Unpacking the boxes, the first thing that caught my eye was the accuracy with which the sample was made. No burr, no sharp bevels, no scratches from the cutter, and there was not even a mountain of unnecessary waste paper with a large colorful box.
The water block has a unique industrial design, which is emphasized by the smooth lines of brushed rough aluminum. The color depends on the version of the waterblock you have chosen. It is currently available in two versions: silver and black. In the center of the water block there is a transparent acrylic camera window through which you can see the two main advantages of our hero. The first is, of course, microchannels, the density of which is simply amazing.
The microchannels are thinner than the blade of a utility knife, and the IHS coverage is 100%. The size of the cooling structure is 4.14×2.98 mm, while the thickness of the rib is about 0.15 mm, and the width of the microchannel is 0.2 mm. The nickel-plated base is quite thick at 3mm, and the maximum microchannel depth is 2.7mm.
Of course, such a feature negatively affects the hydrodynamic resistance. In order to compensate for the GDS, TechN engineers made a number of optimizations. The micro-channels are cut into the base with a special cutter exit radius, which allows the water flow to exit without significant obstacles (except for water friction against the micro-channels).
The second advantage is the optimization of the input stream.
Water enters the microchannels already at a certain angle, while the conduit itself does not have obvious angles, which the water used to “hit” before entering the microchannels. An interesting feature of the TechN water block is the lack of a differentiated jet plate, which water block manufacturers used to brag about so much. Now its function is performed, in fact, by an acrylic transparent camera. This is by no means a minus, but cost optimization and the rejection of really unnecessary details.
As for the details, the TechN water block turned out to be extremely simple. Acrylic chamber, sealing ring, base and, in fact, the frame in which the backlight is mounted.
It is easy to disassemble, while there is simply no chance to assemble something wrong. And therein lies the beauty of it too. Inexperienced users, often disassembling their water block, assembled it with an offset jet plate, which ultimately affected the performance of the water block.
The last important nuance of any water block is the contact surface of the base with the IHS (processor cover).
The surface has a slightly rough texture. The essence of this idea is quite simple. During installation, excess thermal paste can interfere with metal-to-metal contact. In the case of a TechN water block, excess thermal paste will get into micro-scratches, which will ultimately allow for better metal-to-metal contact. But that’s not all. The base has a certain radius of curvature, which makes it possible to improve the metal-to-metal contact exactly at the intersection of two CCDs.
Thanks to all of the above factors, thermal energy is removed as efficiently as possible from each CCD, no matter if you are using Ryzen 5 3600 or Ryzen 9 3950X.
The kit includes a massive reinforcing plate with a dielectric, four screws with springs and Arctic MX2 thermal paste.
To my surprise, even such a trifle as a backplate is executed flawlessly. There are no sharp edges, the thread will never come off, because bonks are used. The springs are tight and have a wide diameter; during installation, the spring will not fail anywhere or the plastic washer will not fall into the spring (I say hello to EKWB!). A hex wrench is included to tighten the screws. Tighten the screws diagonally to create even pressure on the cold plate on the IHS. It will not work to drag or break something, the installation is simple and, as they say, foolproof. Screw recesses allow for a seamless aesthetic.
All TechN waterblocks are backlit. The connection is 4-pin, which allows you to synchronize with any software from the motherboard manufacturer (like ASUS AURA, MSI Mystic Light, etc.). In the dark, this “treasure” looks very impressive, which is sure to please even a hater of RGB.
Testing was carried out on a system with the following configuration:
- Processor: AMD Ryzen 9 3900XT;
- motherboard: ASUS ROG Crosshair VIII Hero Wi-Fi (UEFI 2103);
- memory: Corsair Vengeance LPX 4133C19;
- graphics card: ASUS ROG Strix AMD Radeon RX 5700XT;
- storage: Gigabyte Aorus NVMe Gen4 SSD 2TB;
- drive: Samsung 960 Pro 512MB;
- power supply: ASUS ROG Thor 1200;
- CPU Cooling: EKWB EK Quantum Velocity;
- CPU Cooler: EKWB EK Quantum Magnitude;
- CPU Cooling: Watercool Heatkiller IV Pro;
- CPU cooling: Optimus Foundation CPU Block;
- CPU Cooling: TechN CPU Waterblock;
- pump: EKWB EK-XRES 140 Revo D5 RGB PWM;
- radiator: Watercool MO-RA3 420;
- fans: Arctic P14 PWM PST;
- flow sensor: Aquacomputer Flow Meter;
- temperature sensors: Aquacomputer Aquaero;
- controller: Aquacomputer Aquaero 6 LT;
- thermal paste: ARCTIC MX-4;
- body: Streacom BC1.1 Benchtable.
Condition number 1. Thanks to Aquaero 6 LT and aquasuite, a hysteresis was set for the Inlet port (24 degrees). The essence of this idea is the constant temperature of the water that will enter the water block. Thanks to this, we get unchanging test conditions that are not affected by room temperature. In the end result, this will give us an extremely small error.
Condition number 2. MO-RA3 420 was used to keep the water temperature. The largest radiator with the greatest heat dissipation capabilities (and the largest hydrodynamic resistance). This tandem was complemented by 18 Arctic fans in Push-Pull mode.
Condition number 3. Constant laminar flow, for this the pump speed was selected individually for each water block to get 1 GPM (gallon per minute). Note that each hose kink or elbow will affect flow, so each user system will have a completely different performance.
Condition number 4. To minimize the error, each test will begin only when the water temperature in the system stabilizes according to a given hysteresis.
Condition number 5. AIDA 64 CPU stress test in FPU mode. This is the only stress test that has minimal load spread (the main reason I moved away from Prime95 with a fixed FFT). The time for one test is 30 minutes. Three repetitions for each water block.
Condition number 6. Tests of water blocks were made in two thermal conditions. The first is a standard boost with a maximum PPT of 142W. The second mode is 170 W, this is overclocking on CCX 4500/4500/4400/4400 MHz from 1.325 V.
Condition number 7. To make the task as difficult as possible for water blocks, I used AMD Ryzen 9 3900XT. Processor with high static leakage currents. These processors heat up more, but allow you to reach higher frequencies at a lower voltage. These are the kind of crystals AMD uses in Zen 3.
Nuance number 1. Optimus Foundation CPU Block received only 187 l / h due to the GDS at 100% pump speed. Which is slightly less than the required 240 l/h. But, fortunately, this flow indicator is enough to consider the data valid.
Nuance number 2. On the EKWB Quantum Magnitude waterblock, the Coldplate – Flat base, sold separately for 35 euros, was used, because the base base was too “humpbacked”. Let me remind you that the heat sources of Zen 2 and Zen 3 are off-center, the correct contact of the water block plays a big role.
First summary test results with a maximum PPT of 142W.
The TechN water block was the favorite, beating potential Heatkiller IV Pro by more than 1.1 degrees. The gap from the EK Quantum Magnitude with a flat base was 1.26 degrees. EK Velocity lost by 2.43 degrees. The reason for this loss is due to the lack of architectural optimization of the water block for Ryzen processors with Zen 2 and Zen 3 microarchitecture.
Since the thermal headroom of the processors remained large, I prepared a more serious task with a 170W PPT.
And here the TechN water block showed itself even better, increasing the lead over all participants in the race. In particular, the lead over the EK Velocity is already an impressive 4.26 degrees.
Also, at the request of some enthusiasts, I tested the effect of flow on the performance of a TechN waterblock.
As it turned out, even half the flow (0.5 GPM = 120 liters per hour) allows the TechN water block to remain in the lead. This is undoubtedly great news for those who have a weak pump or a lot of elements in the circuit that affect the hydrodynamic resistance.
The water block TechN, claiming leadership, really confirmed its right to the crown. The guys from TechN are professionals in their field. Speaking of pros, there are many. The design, the quality of the materials and the attention to detail are commendable. The area of the microchannels completely covers the entire IHS. Ultra-thin channels (no defects, this is really unique) speaks of the most advanced technology that TechN uses. The massive base has a huge inertia, which favorably affects the reduction of temperature peaks. The inlet water flow enters the microchannels at a special angle in order to reduce hydrodynamic resistance, while not compromising the performance of the water block.
The TechN water block is also available to users of Intel processors, both the LGA 1200 and LGA 2066 platforms are supported. And most importantly, they are not inferior in manufacturability.
Another important advantage of the TechN water block is its relatively low cost – 100 euros. The closest competitor in the premium segment is Optimus, which will cost the user 15 euros more, while you still have to take care of a high-quality DDC pump. I wrote about DDC for a reason, because most user systems contain corner fittings, filters, GPU water blocks and other elements that affect the GDS. The D5 will certainly cope with the task, but 100% rpm can affect its resource, and you simply will not get the performance of water blocks due to a weak flow (especially on a GPU water block).
I also can’t help but mention the Heatkiller IV Pro, the water block has been on the market for quite some time and is an unparalleled leader in the performance per dollar category, which is important for those who will build an efficient and economical custom system.