It’s amazing what people can do with a little ingenuity, talent (and apparently in this case) patience and time! 
 
Great work, Nateman!

I just made a slightly nicer video.  I am obviously not a photographer, but you can see the plastic top I used is anything but plastic.  Its impact resistant.  I smash it repeatedly throughout the video.  It's uploading now. 

An evening of fluid dynamics.  Screw the computer simulations, just borrow your kids glitter
 

 

Would never have though of using glitter for a flow analysis. That bottom one is best I think when you turn it on and off.

Thanks!  It defiantly has VERY high flow with no restriction at all.  Plus its VERY durable.  I was beating on it with a screwdriver and all the scuff marks came right off.  If I did that to acrylic, it would have been destroyed. 
 
I wonder why all manufacturers don't use polycarbonate?    I think its a worthwhile investment.

Carbide.  Eats through the stuff, but I still am very conservative with my depth of cuts, speeds and feeds.  Helps to keep the mill trammed when there is an absence of sheer forces, but you already know that.  Now am I reading that correctly and your cutting Fpm?!  I still keep things in IPM.
 
I just noticed I became Omnipotent last night.  Who would of thought a fluid analysis would make me omnipotent :)
 
TOTAL side note... I just purchased some I beams to hold up the estimated 800 lbs concrete table I am making for my linear rail build.  Just one more step closer.  Also I found a machinist level at work.  Might have to borrow that :)

The glitter makes it obvious how much water seeps through the parts in the block that you don't want it too.  I have a plan in place, and if you look at the original drawing you can see blue on the fins which is where I wanted to put o-ring material along the fins to ensure 100% liquid flow through the fins. 

After watching this actual experiment, its obvious that liquid does flow inbetween the top plate and the main division of the block.  Its not exactly a show-stopper since 95% of the water goes where it is supposed to go simply by there being less resistance along that path, but there is some fluid that bleeds over directly into the exit port.  I am designing an additional O ring channel to guide the water along its intended path.  Just brainstorming at the moment. 

nateman_doo

Carbide.  Eats through the stuff, but I still am very conservative with my depth of cuts, speeds and feeds.  Helps to keep the mill trammed when there is an absence of sheer forces, but you already know that.  Now am I reading that correctly and your cutting Fpm?!  I still keep things in IPM.

I just noticed I became Omnipotent last night.  Who would of thought a fluid analysis would make me omnipotent :)

TOTAL side note... I just purchased some I beams to hold up the estimated 800 lbs concrete table I am making for my linear rail build.  Just one more step closer.  Also I found a machinist level at work.  Might have to borrow that :)

lol yah you got me there. i meant IPM. (brain fart) I wish my machine would go 3600ipm:) some do btw. but still, its a huge diferance when i can do 1 pass at 300ipm on acrylic compared to 2 passes at 50ipm on polycarb:)

Thanks!  nope on the Gibbs, I use ProE Wildfire for the CAD.  The CAM is pretty recent, its EnRoute4. It is a pretty limited program,  but it gets most of the legwork done as far as tool paths.  I am becoming more and more familiar with it after Gratuitous spent about 4 hours talking me through the program.  It has lots of useless tool retracts after it does an outer profile (which added over an HOUR to the overall machining of this block), so knowing the G-code helps so I just manually manipulate the tool paths. 
 
Only problem is the human factor.  If I want a 0.05 depth of cut (usually I am 0.02 on my tiny X2 mill), if I fudge the code and put Z -0.5 instead of 0.05... I am going to know I screwed up very quickly as I destroy a carbide bit and throw the machine all out of tram.
 
24x24x8 is a VERY good sized work envelope for the machine your planning.  The one i am building is 18x18x18 simply because the linear rails are 24" NSK's.  If I bought longer rails then I could have made something bigger.  Then I just have to ask myself, going from 11.5 x 7.5 x 12 to 18x18x18, what do I really need such a large work envelope for?
 
I am sure one day I will wish it was something bigger, but for what I do here now with making waterblocks (and whatever odd jobs) the Y axis was always the limiting factor.  I will be able to drop a 12x12 sheet of copper on it, and mill out both top and bottom of a waterblock (at least thats the idea).  And more importantly, run a 3" fly cutter across an entire surface of a 12x12 sheet of stock.
 
The raw material I have is usually ok, but there is always a huge gash somewhere from the goons in the warehouse manhandling the material, or bad packaging.  I try to compensate by positioning the gash into a water channel or someplace that won't effect the outcome.
 
What type of spindle box/prefab head do you plan on using?  Or you going all out and making one from scratch?  Do you already have access to a CNC, or a manual bridgeport type?

THK and NSK seem to be some of the best (and affordable) linear rails I could find.  I just went with NSK because they were available.  If you have the time to build a new head then go for it.  I am not at that level yet, but hopefully one day I can be.  I bought a Tormach V770 head for this new build.  If only I had the time to work on that build, I still have another CNC that I have to make. 

Thanks.  I now have 70 LBS OF COPPER (31.6KG for you metric types) that will soon become FOUR waterblocks!
 
I am not even close to exaggerating, 70 lbs of copper.

I have a slab of copper already on the machine, but I have a busy weekend.

Sorry for the delay, a bit under the weather today.

Any idea of what the final weight will be for the waterblock?  From the pics you provided, it looks like it'll be fairly heavy.

Will this cooler require some extra support to relieve the strain on the video card, such as a back plate?
 
Seeing images like this really makes me wish that I would've stuck with being a machinist to indulge in projects like these.  Although if I did, then I would've never got into computers.