• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

Alphacool Eisblock ES Acetal RTX 3080/3090 Reference

VSG

Editor, Reviews & News
Staff member
Joined
Jul 1, 2014
Messages
3,726 (0.94/day)
Alphacool has two very different blocks for the RTX 3080/3090 reference PCB to cater to different tastes. This article goes over the ES Acetal, a non-flashy block with side-mounted ports, an included backplate, and a unique coolant layout for a decent price.

Show full review
 
Great review, I've got this on my 3090 and it's pretty damn good. Installation was simple as anything and temperatures are less than half. Radiator and pump setup I'm maybe a bit over the top with the eiscool external radiator setup.
 
  • Like
Reactions: VSG
Thanks.
 
  • Like
Reactions: VSG
Took me a couple of minutes to get my head around the water flow path caused by the end-of-card inlets.

It seems to cool half of the first VRM bank on the way in, then a quarter of the flow is diverted for all of the other VRM bank before rejoining at the GPU/VRAM coldplate, and then cools the other half of the first VRM bank on the way out?

Given how much machining and effort went into routing out two-layer, multi O-ring channels for the VRMs, I'm really quite surprised that the area of liquid contact of those VRM plates is both tiny and polished smooth. It seems like a waste of all that effort to get LOTS of water at a high-flow/low-impedence past those VRMs and then barely make the effort to do anything with it.

Perhaps this is just a side effect of very limited clearance around those VRMs indeed - I guess the clearance is so ridiculously tight that it mandated this little notch in one of the VRM plates:

1623790414209.png
 
Took me a couple of minutes to get my head around the water flow path caused by the end-of-card inlets.

It seems to cool half of the first VRM bank on the way in, then a quarter of the flow is diverted for all of the other VRM bank before rejoining at the GPU/VRAM coldplate, and then cools the other half of the first VRM bank on the way out?

Given how much machining and effort went into routing out two-layer, multi O-ring channels for the VRMs, I'm really quite surprised that the area of liquid contact of those VRM plates is both tiny and polished smooth. It seems like a waste of all that effort to get LOTS of water at a high-flow/low-impedence past those VRMs and then barely make the effort to do anything with it.

Perhaps this is just a side effect of very limited clearance around those VRMs indeed - I guess the clearance is so ridiculously tight that it mandated this little notch in one of the VRM plates:

View attachment 204059
Always had my doubts about these tiny VRM "plates". Without much in the way of transfer, they may very well end up being sinks.
 
You don't need much contact area for VRMs, the actual components are tiny. It's been the same for motherboard VRMs in VRM blocks and monoblocks too. The residence time for the coolant as well as the thickness/conductivity of the materials between the VRMs and the coolant are waaaaaaay more important.
 
Back
Top