Mk.6 EMP -Experimental Modular Project-

Showing my ignorance here but what the heck,here goes.....what is the clear/frosted perspex thing in this shot?:


A key to tighten/loosen the black blanking piece/stopper?

 

Great OC build/mod, looks like a G5 case, is that on purpose or has it just "evolved" into that...

 

i do believe thats exactly it

 

Yeah. The cap cannot be tightened by hand to compress the O-ring far enough to make a seal. And since the fillport cap is a made-up bit of hardware that nobody makes wrenches for I made my own in a manner of minutes using the band saw.

Someone yesterday said the case looked like a Decepticon version of the G5 case.
It's not at all intentional because at first I was trying to design a rackmount case like the one I have now. They come with handles on the front, which is fine for pulling them out of racks but useless for carrying them around because PCs are very rear-heavy.
But I couldn't keep the machine within rackmount specification and still provide ample room for watercooling equipment and the docking system I came up with so the case became larger.

Atleast It's being compared to a nice looking case (albeit a proprietary one though) and not something ugly like the Asus Vento.

And it's mostly just the handles that leads to the comparison. The inside is actually quite similar to the Cooler Master Stacker case, but with a few more of the parts being removable and the case frame itself can be taken apart.

 

Just did a leak test with the fillport and the compression cuff and YAY! They both work (as in they don't leak when subjected to 30PSI).

 

Like all of your projects I'm a fan of this one too. Keep up the awesome work!

 

My body has been pretty miserable for the past 2 days because of the weird weather here in the Northeast. I pretty much slept all day wednesday because moving was too uncomfortable.
I have however finished the CAD models for both drive cages.

And I started work on modifying the fan controller and I finished countersinking some screw holes on the case frame and various parts to make them fit together much tighter.

If I can move on Thursday I'll go back to the machine shop. If not I'll just keep working on a few things at home.

 

Halfway through milling the base of the CPU waterblock.

And I'm working on the reservoir sides.

 

Wow, first time I've seen a block with that design on a bit of copper. Othervise poeple only thinking 'bout doing those kinda blocks. Good luck to you mate, cant wait till the nest update =)

 

Everytime I see an update to this project log I am immensly impressed by the skill and work; looking forward to seeing finished project.

 

That is the most amazing waterblock ive ever seen. I cant wait to see the rest of it, and it it works as well as you intend.

 

Absulotely agree.

 

first off all, sorry to hear about the bodily problems, atleast your working around them, and i hope you get well enough to continue modding

second, this project is truely incredible, the fact that your doing nearly all of it by yourself, building from scratch, and still getting good results is always good to see, and the fact that your not settling for something less than par, as in, youve seen your goal, your going for it, i really like to see that in a project.

the waterblock looks incredible, and the amount of surface area, should see it performing very very well.

makes me want to start my multimedia rig.

keep up the incredible work! and i cant wait to see more progress!

 

it looks amazing, but i dont quite undertand how it will work, the channels all go to the outside, but they dont join with the outside, i dont understand how the water will flow through it - time will tell ey

 

look at the acrylic lid hes made, the outside channels all join in the acrylic, which then goto an outer barb, simple.

the inner barb has multiple jets to send the water down the individual channels.

 

hmm, i can kinda see what your talking about
all will be made clear soon enough i hope also show some pics when assembling it pls i wanna see how it goes together (if u dont mind )

 

I'm very flattered. I have seen far more ambitious and scientifically researched designs, but what I've come up with owes much to the numerous individuals who have been kind enough to give me honest objective criticism to help ween me towards more effective configurations. I'm designing around what tools I have access to and what numbers and configurations are known to work with only minor revisions based on my limited understanding of the principles involved. Imperical evaluation once the block is done will point to whether or not my assumptions were correct.

Thank You. I always work under the belief that the most beautiful things are those that were designed to accomplish a functional goal, but in a gracefully effective manner. This project is 90% about function, and in being so I have very exacting desires for what I want the end product to be and do. I wouldn't feel right out-sourcing any of the work when I'm perfectly capable of accomplishing it myself. Regardless of whether or not it's difficult; if I can make it with my own hands I will.

The waterblock will make much more sense in the next week. The block is really just a sandwhich of three pieces. The top, the nozzle plate, and the base.

The center inlet that goes through the top widens and is then divided into 26 (I think, didn't bother to count) nozzles in the nozzle plate. The nozzles cover the square inch in the center of the base.
The flow then spreads out radially from the center and the 2 45-deg bends in each channel are placed at the approximate edges of the IHS of the Opteron 148 I will be cooling. This means that the entire processor contact area on the base is receiving fairly turbulent (and therefore thermally efficient) waterflow.

The outflow from the base is reconvened towards the outlet through the channel in the top. The machining of the top is not very complicated thanks to the fact that I can simply adhere the nozzle plate to it.
Thanks to ALOT of pre-evaluation and planning in Solid CAD the only time consuming part of the block is the insane amount of work involved in milling the base. I already have 5 hours put into it and it looks as though I have 5 or 6 more to go.

Each straight channel length you see in that photograph requires 3 passes because a 1/16th" bit can't cut out anymore depth than .05" without breaking. Even at the super-slow feeding rate I'm working with.

I'll try to be more detailed with my photography as the project has started to get too complicated to just have "end-product" snaps that don't really explain what the hell it is I just made.

 

The reservoir is underway.

And the GPU and Northbridge blocks are installed.


And this is the revised waterblock design I'm still in the process of machining.

 

So, what exactly did you need the forceps for in this? Or were they just lying around?

And, uh, huzzah for a cool looking resevoir! Wish I had something more intelligent to say...

 

how many hours have you put into the EMP so far?