Improved heatsink
Let’s now look at the new model with the large heatsink, which I had so eagerly wished for since the first test in this form. You could already see in the first pictures that the heatpipes were also modified to allow more air towards the PCB and VRM cooler. Bending in particular has been optimized once again. We’ll talk about the significantly larger heatsink in a moment, because that’s where the real innovation is. Of course, the weight also increases a bit and 575 grams become a generous 620 grams, but that is still significantly less than most mid-range graphics cards. And the direct comparison coolers all weigh well over 1000 grams.
The first thing we see is that the memory is now also actively cooled. i had spoken out against a sandwich solution of GPU copper heatsink and light metal frame, as many cooler manufacturers unfortunately like to use for cost reasons. This one is solidly milled from one piece. Moreover, we also see that Raijintek has managed to keep the original bracket solution. I’ll get to that in a moment, too.
We can see very nicely in the picture of the prototype below how the neatly bent copper heatpipes (inside it is a composite material) were tinned and then soldered to the copper heatsink. Since it is a manually produced prototype, the solder joints are so easy to see. In the final version, you can no longer see all this under the nickel coating. But so I could show you the nice times and you know at least what’s underneath. And you may see six 6-mm heatpipes on superficial inspection, but there are actually as many as twelve!
Unfortunately, I can not break off the heatsink, but it is already a kind of well-kept secret that Raijintek does not lead the heat pipes closed through the heatsink, but simply divided them in the middle of the heatsink. Well, the knowledge is one thing, but the technical implementation is then much harder than you think. That’s exactly why you don’t see it so often. The reason is simple, but expensive. A heat pipe only really works optimally in one direction. However, if you heat it up in the middle and hope that the transport of the waste heat to the two cooler ends will also work well, then you have made a complete mistake with an asymmetrical design. Yes, it still works, but not optimally. Unfortunately, using two individual pipes instead of one continuous one also costs almost twice as much because significantly more work steps and precision are required.
I already mentioned that the brackets and spacers are new. They are quite easy to screw from the top. The M3 screws intentionally first engage in the base plate of the heatsinks on the back of the heatpipes and not in the upper part where RAM and GPU rest. This takes partial pressure off the heatsink and achieves a more even support. It also presses the heatpipes up a bit against the nickel-plated copper heatsink after screwing them in, which is desirable. Yes, it’s also the small details that can boost cooling performance a bit more. That’s certainly not much for any innovation, but in the aggregate you do notice it. And that was exactly the goal.
Finally, you simply place the PCB upside down on the cooler and screw the whole construct with four 2.5 mm spring screws. Ready!
But stop, we still need air movement. That comes now on the next and thus penultimate page.
104 Antworten
Kommentar
Lade neue Kommentare
Veteran
Veteran
Veteran
Mitglied
Urgestein
Urgestein
Urgestein
Urgestein
Veteran
Veteran
Mitglied
Urgestein
Urgestein
Mitglied
Veteran
Mitglied
Urgestein
Urgestein
Urgestein
Alle Kommentare lesen unter igor´sLAB Community →