Bustin' out a Q6600 in a small cube clear acrylic case - assembly pictorial

[pclausen]

I have never done a water cooled rig before but I decided to jump in with both feet and doing one up in one of those Sunbeam clear acrylic cube cases (the UFO).

The parts I'm going to stuff into this small cube are as follows:

Asus P5E-VM HDMI
Q6600
2 x Crucial Ballistix DDR2-800
2 x Raptor 150s (configured as a small "enter the Matrix" raid0 boot partition, and a larger raid0 data partition)
LG Hd-DVD/Blu-Ray drive
Corsair 520HX

The cooling system is as follows (also all stuffed inside the cube except for the radiator):

MCP655 Pump
MCRES-MICRO reservoir
Apogee GTX
MCR220
2 x Yate Loon D12SH

I'm expecting the Asus tomorrow (it just got released last week), but while I wait, I went ahead and mounted the radiator and loosley fitted the other items:

Side angle (here you can see the reservoir in the front upper left corner):

http://www.cstone.net/~dk/q6600angle.jpg

Top (MC655 visible on the "lower level" and of course the CPU block on the "upper level"):

http://www.cstone.net/~dk/q6600top.jpg

Front (here you can clearly see the "upper level" where the mobo goes, as well as the lower level containing the drives, PS and pump):

http://www.cstone.net/~dk/q6600front.jpg

Radiator:

http://www.cstone.net/~dk/q6600radiator.jpg

The fans are about 1.5" away from the back of the case. There is no hole into the case since the PS takes up about 3/4 of that area anyway. I'm a little concerned that the Yate Loon's are going to be restricted a little, but perhaps not?

I need to pick up a 3/4" hole saw to drill the 2 holes for the 1/2" I.D. tubing to enter the case from the radiator. Since the Asus is a Micro-ATX, there are enough vent holes to allow me to run the rest of the tubing with no additional drilling. So the flow will be as follows:

Lower reservoir exit -> pump inlet
pump outlet -> lower radiator inlet
upper radiator outlet -> cpu block
cpu block -> upper reservoir inlet

I only expect to end up using like 3 feet of tubing being how compact everything is. I'll likely also go through the pain of cutting all the electrical wires to exact length and solder and heat shrink them. The Corsair is modular, so I won't kill it by doing so.

I plan to run the 2 chassis fans on the upper level from the "chassis fan" header on the mobo. Likewise, I was planning on running the 2 Yate loons from the "cpu fan" header on the chassis. This way the chassis fans will only speed up enough to maintain a certain temp on the upper level, and the radiator fans will likewise only run fast enough to keep the cpu cool.

I do have the "slackr" stepping, so I'm hoping to at least get 3.6 out of it. This machine will mostly be for playing Blu-Ray and HD-DVD, but will also do a fair amount of ripping and transcoding. No gaming.

Anyway, like I said, this is my first attempt at water cooling, so please speak up if you see me already doing something obviously wrong!

I'll post more pics as the assembly progesss continues.

[Jas]

Should be a neat build when you done, looking good!

[Clockwork_Apple]

Looks like it's gonna be a good build.

sub'd :)

[voigts]

For cutting those holes for the tubing, DON'T use a hole saw. They heat up the acrylic too much and make a mess. Get a simple wood boring spade bit like this:

http://www.whitecapdirect.com/store/images/products/173-88804.jpg

The bit sheds off the material as it cuts and won't heat up the acrylic.

The trick to a case like this is to do some serious sleeving/cable management because everything shows. I haven't seen this case before. It looks pretty neat although it also looks like it is going to take some work to get everything stuffed in it neatly.

[pclausen]

Thanks for the tip. I'll pick up a fresh spade bit then as I think the ones I have laying around the house are quite dull. A 13/16" or a 7/8" should work well I would think as 3/4" will likely be too tight since that is the actual O.D. of the tubing.

[Jas]

I can't wait to see what you come up with!

It's going to be interesting to see how it all comes together in that small case!

[pclausen]

Man, UPS sucks around Christmas time. My mobo didn't show. Last scan is from yesterday. I called them as they told me to ask the shipper to put a tracer on it tomorrow if it is still not in the system by them.

Anyway, I did manage to pick up a 13/16" spade bit and it worked like a champ. I was also able to permanently mount the pump tonight.

Here's a shot of the hose going from the pump outlet to the lower inlet on the radiator (don't worry about that upper radiator hose, its just temporary):

http://www.cstone.net/~dk/q6600hoses1.jpg

And a shot from above showing the hose going from the reservoir (still not mounted until I get the mobo) to the pump inlet. Notice how the hose fit nicely through the vent from the upper deck to the lower one:

http://www.cstone.net/~dk/q6600hoses2.jpg

[yellowhello]

Looks awesome! I recently just got one of these: http://www.mountainmods.com/h2go-mirror-black-powder-coat-aluminum-cube-computer-case-p-420.html

[sunrunner20]

in my experience theres a bit thats even better than the spade for cutting plastic... The Forstner bit! Kinda pricey though.

No matter what though... Take it SLOW and with light pressure. Also, go to your local hardware store and buy up some of their scrap plexi. Practice on this before drilling into an expensive case.

PS: I have a drill press so a Forstner might not be the best choice if you don't.

[Sam__]

looks like its gunna be a good'n...good luck

[SuperDave1685]

If I was you man, I'd place those yate loons on the other side of the Radiator, pulling air through. On the side closest to the case, you're not giving them a whole lot of breathing room, regarldess of whether they're pushing or pulling.

[pclausen]

Sunrunner, I do have a drill press and things did work out well with the spade bit. You're right about going real nice and slow through! I drilled small pilot holes initially as well.

Superdave, I hear you on the tight space for those yate loons. I'll look into moving them to the other side of the radiator. Long term, I'll likely switch from using 2" 8-32 screws to hold them to some sort of beefier spacers that are longer. I'll likely wait until I get the rig up and running to see what sort of temps I get out the gate.

[Captain Slug]

For any holes larger than 3/16" or 1/4" I've had the most luck with spade bits, hence why I've recommended them so much. I'm also rather fond of brad point "lip and spur" bits which produce very clean holes in plastics.
Going down to as slow of a speed as you can gear your drillpress to helps a good deal. Also use very minimal down force when drilling. Keeping the material as close to room temperature as possible when working with it is always advised for keeping control of the piece your working on and for minimizing fume production.

I pretty much avoid acrylic now. It's just too annoying to work with for most applications.

[pclausen]

My Asus mobo finally showed up today so I was able to complete the install. I haven't cleaned up the wiring yet, but I wanted to get this sucker fired up. So here are some pics of the rig up and running:

http://www.cstone.net/~dk/q6600water1.jpg

http://www.cstone.net/~dk/q6600water2.jpg

http://www.cstone.net/~dk/q6600water3.jpg

http://www.cstone.net/~dk/q6600water4.jpg

http://www.cstone.net/~dk/q6600water5.jpg

http://www.cstone.net/~dk/q6600water6.jpg

I just tilted it back and fourth a little and that seemed to get out the bubbles/air. I suspect it will take a few hours for all the microbubbles to work their way out. But so far the cores are only 17 degrees C according to the BIOS!

[Vagabond102]

Nice build!

Did you leak test on live equipment? *shudder* :-)

[Moto7451]

Looks really good. Usually acrylic cases don't appeal to me but yours came out well.

It's ok to leak test with your stuff in the case. You just need to run the pump off of a separate PSU. Also I'm pretty sure his comments about tilting the case back and forth was after he leak tested ;).

[samuknow]

I just built one of these for a review. I am in the process of WCing it. Very nice layout. I like the rad at the bottom. I was thinking of putting it on the top fan. I am still not sure how to coll my video card. I guess i will run the tubes out the top....Nice job.

[pclausen]

Nice build!

Did you leak test on live equipment? *shudder* :-)
Well, I added as much distilled water/coolant mix as I could while tilting the unit, I then went ahead and fired up the rig and kept adding water as needed until my reservoir was above the fill mark. In hindsight, using an external PS just for the pump at this critical stage would certainly have been safer!

I have done about a half dozen cooldowns and just began running prime @ 3.4GHz. Everything appears stable and temps are only just approaching 40c. I feel getting to 3.6 or even 4 should be possible. This new BIOS has a lot of OC settings I haven't seen before, so there is lots of stuff to play with.

http://www.cstone.net/~dk/prime1.JPG

[SuperDave1685]

Wow.. Great temps!! Makes me want to get a G0 CPU :(

[Jim Morbid]

Thats a really great looking build!

Congrats are in order pclausen.

JM

[voigts]

Looks great. So far a very good job in such a transparent case.

BTW, since you have so much acrylic, I would suggest picking up some Novus #1 plastic clean and shine. I use this as I have a lot of acrylic in my setup and this stuff really works. It shines and keeps the dust from clinging to the acrylic as much. I bought it and wondered if it was good or not, but it really does work well.

The Novus #2 works well on minor scratches also.

[Jas]

Looks great!

That CPU look sliek it takes the voltag enicely. Wish mine wasn't such a voltage hog!

[4od]

Wow, i didn't know the Apogee GTX looked so awesome!

Nice setup, now go for 4GHZ!

[pclausen]

BTW, since you have so much acrylic, I would suggest picking up some Novus #1 plastic clean and shine.

The Novus #2 works well on minor scratches also.
Thanks for the tip. I was wondering what would work well to keep it clean. Where can I pick up that stuff?

Nice setup, now go for 4GHZ!
Hehe, I just passed 9x400 stable (3.6GHz) and my cores are all in the range 46-49c. I have yet to see a delta of more than 3 degrees between any 2 cores at load.

To get to 3.6GHz, my BIOS CPU voltage is 1.5625, but if I'm understanding correctly, the voltage that matters is the one being reported by CPUz under load, and that voltage is still only 1.360V, so that means I still have 0.140V to play with to get to 4.0GHz, right?

[voigts]

Thanks for the tip. I was wondering what would work well to keep it clean. Where can I pick up that stuff?

I googled it and the #1 ranking is the best price I can find.

http://www.delviesplastics.com/waxes_cleaners_novus.htm

I bought some off of Ebay when I had some acrylic laser cut a while back.

To get to 3.6GHz, my BIOS CPU voltage is 1.5625, but if I'm understanding correctly, the voltage that matters is the one being reported by CPUz under load, and that voltage is still only 1.360V, so that means I still have 0.140V to play with to get to 4.0GHz, right?

Although not new at watercooling, I finally decided to get off my butt and put all of this expensive cooling to work and figure out how to overclock. So I would like to know the same thing as I just worked on figuring out the max my RAM will do today at 2.1v (996mhz-not bad). Now I also want to see if I can get my 6400 to 4ghz.

[pclausen]

Thanks for the link voigts! I ordered up a bottle of each.

Still a little concerned about the voltage thing. CPUz is reporting 1.508 or something like that at idle and then drops to 1.360 under load. I believe Intel rates the Q6600 for 1.5V max, but I understand it will take up to 1.6V before permanent damage would take place.

A couple of "night pics":

http://www.cstone.net/~dk/q6600water7.jpg

http://www.cstone.net/~dk/q6600water8.jpg

[Mycobacteria]

Is the PC on its side

Or is the pump side mounted and is this ok for the pump?

[Jas]

Pump is mounted on it's side, and thats fine for a pump

[Jas]

To get to 3.6GHz, my BIOS CPU voltage is 1.5625, but if I'm understanding correctly, the voltage that matters is the one being reported by CPUz under load, and that voltage is still only 1.360V, so that means I still have 0.140V to play with to get to 4.0GHz, right?


Rather than repeat all this crap, check out this sticky, (http://www.ocforums.com/showthread.php?t=515316) that should help you guys out.

Depending on which version of CPUz you ar eusing it may be reporting too low. My E6400 is taking 1.52v to hit 3.6 stable as reported in CPUz.

[voigts]

Rather than repeat all this crap, check out this sticky, (http://www.ocforums.com/showthread.php?t=515316) that should help you guys out.

Depending on which version of CPUz you ar eusing it may be reporting too low. My E6400 is taking 1.52v to hit 3.6 stable as reported in CPUz.

I don't see the voltage question about the difference between BIOS voltage and CPU-Z and hence what limit to go by mentioned.

[David_N]

Wow, looks fantastic!

Subscribed for further pics and my own reference later :)


Great job.

[Jas]

I don't see the voltage question about the difference between BIOS voltage and CPU-Z and hence what limit to go by mentioned.

There was a parapgraph in there about the different versiosn of CPU reporting the Vcor too low. And also the VDrop on mothe rboards.

To sum up, CPUz versions less than 1.41 is probably reporting too low of voltage. If yo uhave 1.41 then it should be accurate. Under high load you will get a Vdrop in your voltage, so what you want to look at is your voltage with load, and without load. See what your Vdrop is.

[samuknow]

I have to say again....looks great.

I do hate you still for beating me to this.....LOL


j/k