This is a mirror of my primary design log, which is on blogger
Since my first water cooled computer in 2004, I've been unhappy with the computer cases which are available. It's fairly simple: either the radiators are very obtrusive, the case is very cluttered or the most elegant parts of the build are hidden from sight. In recent times, I've seen a few builds which solve all of these problems with some heavy modifications to the original case. I decided a while ago that if I were going to have to make such massive changes, I might as well just start from scratch.
I like working with metal and with wood, but I have far readier access to woodworking equipment than I do metalworking equipment and I find wood more forgiving, so I'll make this case primarily from wood. The combination of wood and water cooling led me to consider doing a steampunk build. I think that steampunk done well can have a really good aesthetic--see this amazing build for an example of how beautiful well executed steampunk can be. On the other hand, just pasting gears to everything not the kind of aesthetic I want.
This build will go together in a few stages, since several of the points are going to require me to design some intense electronics. I'm doing the design in ske
tchup and, over the next few weeks, I'll be putting my design up here as I get it together.
The specifications of the computer aren't here because this is going to be a long-running project--I likely won't buy the actual computer hardware until 8-10 months from now.
Here's a rough sketch of the overall configuration; where components will go, etc.
Very rough. I already know that the drive bays are the wrong size.
The water cooling in this build will be on every component which might even fractionally benefit from it. I plan water cool the CPU(s), GPU(s), RAM, chipset, and MOSFETS. The disk drives will be in a separate chamber, and will not be water cooled. One unique feature of this build is the piping: I intend to do virtually all my tubing in copper pipe.
I'm still undecided on a few important points, such as whether to build, scrounge, or purchase the drive bays.
Stay tuned, I'll be updating with component designs as I go.