Crawl Space Cooler Build

[bigslappy]

[QUOTE=mike8913;2969357]it depends on the volume of the fall. For it to work properly there needs to be more volume of water being pushed down by gravity than the volumne of water that needs to be pushed back up.[/QUOTE

Yes !

i'm talking the two loops flow out of the PC into a 8 inch pipe just at the level of the PC then a straight drop of 7 feet (5 feet of that drop 8" pipe) reducing down in diameter(THE LAST 2 feet of the drop ) every 3 to 4 inches till the diameter is a 1 inch pipe that tee's at that point back into the two loops
then both loops run into a rad , one on each loop (Black Ice GTX-480's) & is pushed back up by a iwaki rd-30 pump on each loop back into the PC

The volume of water in a 8" pipe 5 feet long straight down will supply alot of positive pressure to the inlet of the rads & the RD-30 will create a negative pressure / suction at the outlet of the rad
the up loops will be the diameter of the outlet of the Iwaki's
the pipe between the rad & Iwaki will copper pipe to keep the Iwaki from crushing the tubing with suction
does that make sense ? Opinions are welcome !

[bigslappy]

I found a heating / ac contractor that has a sheet metal shop that makes duct work about the same height / width as my wood box ..SOO if this works then i'll move on to that instead of the woodbox & also use a brand new blower but only after this thing is up & running correctly

QUESTION .. should i put the the Iwaki's before the rads ?

or as I was thinking after the rads for the push up to the PC

[mike8913]

pump placement shouldn't matter. Some people suggest after the rad for velocity but common sense would dictate that the same force being pressured out the pump has to also be sucked into the pump. And, since your pump has quite a large heat load i would personally suggest putting in before the rad. Now granted, even though common sense does dictate the previously stated it could still be that there is a performance drop in that alignment. However, I hardly think that any performance drop (if there is one) would be noticeable.

[Giannis86]

very interesting build you have there
i would put the iwaki before the rad to account for the larger heat dump

*edit: mike beat me to it xD

[desertstalker]

See http://www.lerc.nasa.gov/WWW/K-12/Wi..._pressure.html

The Pressure of a fluid is determined by density and depth, the volume has nothing to do with it.

This is because a fluid exerts pressure equally in all directions.

Or are you going to make be dig up my Fluid Mechanics notes to find the derivation for the formula : P = Pa + roe*g*h (notice the absence of any dimension other than height)

I believe you may be confusing force and pressure.

[bigslappy]

See http://www.lerc.nasa.gov/WWW/K-12/Wi..._pressure.html

The Pressure of a fluid is determined by density and depth, the volume has nothing to do with it.

This is because a fluid exerts pressure equally in all directions.

Or are you going to make be dig up my Fluid Mechanics notes to find the derivation for the formula : P = Pa + roe*g*h (notice the absence of any dimension other than height)

I believe you may be confusing force and pressure.

in other words you're saying i'm wasting my effort placing a large volume to use as a down ward pressure to the rads ?, if that is the case then i would want the Iwaki's above the floor so i could monitor them as well as the PSU's that supply the 24Volts to the Iwaki's ?


Then how about this Two loops in PC > Iwaki's > down to the rad's /blower > up to the resevoir's > PC

with this setup the heat dump of the Iwaki's is before the rads & after the heat dump of the PC water blocks >

>> the Iwaki's should have enuff headroom to pull this off ?

[H2omg]

See http://www.lerc.nasa.gov/WWW/K-12/Wi..._pressure.html

The Pressure of a fluid is determined by density and depth, the volume has nothing to do with it.

This is because a fluid exerts pressure equally in all directions.

Or are you going to make be dig up my Fluid Mechanics notes to find the derivation for the formula : P = Pa + roe*g*h (notice the absence of any dimension other than height)

I believe you may be confusing force and pressure.

I had a feeling was I was right, but just didn't know enough about it to defend my thinking. I've always had a good understanding of the physical world, but don't know any of the math to back it up.

[desertstalker]

In theory you should place the pumps intake near (at the same level or below) the RES, they run better with less restriction on the intake and it will be easier to prime them.
It will make very little difference where you put the components as there will likely only be a small temperature increase through the whole loop (as water has a high specific heat)

I expect the Iwaki's will have plenty of power for what you want.

You may want to run it; RES -> PUMP -> RADS -> System -> RES but it makes little difference.
Just make sure you have the pumps below the RES so you can fill the system.

[eternaljammer]

Absolute beast of a project! Thanks slappy for building this unit. I will be monitoring this thread!

[F1ZZY]

See http://www.lerc.nasa.gov/WWW/K-12/Wi..._pressure.html

The Pressure of a fluid is determined by density and depth, the volume has nothing to do with it.

This is because a fluid exerts pressure equally in all directions.

Or are you going to make be dig up my Fluid Mechanics notes to find the derivation for the formula : P = Pa + roe*g*h (notice the absence of any dimension other than height)

I believe you may be confusing force and pressure.

Isn't pressure just the amount of force per unit area?

I think what people are trying to explain here is the potential energy contained in the column of water due the the gravity effecting it and its weight, so the thinking is big volume is more energy down.

I don't think this will stand true as it is a closed loop. No water is able to go down until it returns to the res. (finger on straw = lift from drink) 1 cc of water pumped in to the res = 1cc allowed out of the res. The res water will be "held up" with surface tension, ie defeating gravity.

I think there will be a point where it all gets messy with large enough and long enough res's that enable the surface tension not to be strong enough.

For all intents and purposes though, res volume works as a buffer for temps and not much else. Correct me if I'm wrong here. I'm getting old and rusty.

[H2omg]

Isn't pressure just the amount of force per unit area?

I think what people are trying to explain here is the potential energy contained in the column of water due the the gravity effecting it and its weight, so the thinking is big volume is more energy down.

I don't think this will stand true as it is a closed loop. No water is able to go down until it returns to the res. (finger on straw = lift from drink) 1 cc of water pumped in to the res = 1cc allowed out of the res. The res water will be "held up" with surface tension, ie defeating gravity.

I think there will be a point where it all gets messy with large enough and long enough res's that enable the surface tension not to be strong enough.

For all intents and purposes though, res volume works as a buffer for temps and not much else. Correct me if I'm wrong here. I'm getting old and rusty.

My understanding is the same. The more water in the loop the longer it takes to saturate. The guys here refer to it as "heat dump" I believe.

[bigslappy]

1st off thanx to you ALL for all the feed back it has helped Very Much ! & thanx also to Quoc from Petra's Tech Shop for all the help (those guy's are great!)

i'm not trying to re-invent the wheel here & i'm no expert on fluid mechanics , i just know that I got a huge volume of cool stable air just under my feet & i get free power from a solar photo array on the roof ... soo why not use the free cool air ,free power to my advantage (solar units in place now 6 yaers & i have hit the break even point on costs for them )

I'm going with a set-up thus ... pc > Rad (GTX480 BlkIce){this will be the under floor/crawl space blower unit } > Res-(EK-RES400 Rev.2) > PUMP-( Iwaki RD-30 )> PC

2 loops of each .. loops having it's own rad /pump/res

the reseviors,pumps & PSU's for the pumps will be housed in a sound box much like CyberDriud's build with this box on the floor with the rad/blower directly below about a foot (the width of the floor/subfloor/joists--- with the pc directly above both units at desk level
up/down is about 6 feet or less each way .. blower/rad unit to PC

I'm on to testing /flushing rads right now, will use hot tap water straight outta the hot water heater > garden hose > rad will take a reading of water temps at the inlet of rad & another reading of temps at the outlet of the rads all the while with the blower unit running to see what temp drop i get
pix to follow in a few hours

[goramus]

couldnt you add a Y to the big down pvc tube to add a air vent so you can get the down force your looking for? it wouldnt be a closed loop any more, could add some sort of air filter to it so you wouldnt get gunk in your lines. either way tho you sure do have a really cool idea and build going on. cant wait to see pics of everything when your done!

[F1ZZY]

That's a novel idea, but as long as there is a return water pipe, height is everything and volume means nothing. (while static at least)

OK, I have thought of a way to simplify explaining this.

Imagine a simple U shaped pipe with one arm twice the diameter as the other. The water would still settle evenly in both, regardless of water volume.

But potentially it could be easier to lift the water up the thin tube than the thick one. This is making my head hurt!!

Too be honest, with the amount of air and water flow available, I wouldn't really worry about res design at all.

[H2omg]

Imagine a simple U shaped pipe with one arm twice the diameter as the other. The water would still settle evenly in both, regardless of water volume.

But potentially it could be easier to lift the water up the thin tube than the thick one. This is making my head hurt!!

That's an interesting explanation, really gets the mind thinking. If you were to put your mouth on either end and blow, I imagine it would be "easier" to blow the pipe empty from the larger diameter end. All this theorycraft makes me wish I had ambition and a college education lol. : (

[bigslappy]

temp test went as follows ; Blower on at all times
rad #1 house hot water inlet temp 120F outlet temp 98F air temp 72F

rad #2 inlet temp 120F outlet temp 102 air temp 75F

then was dis-assembling the test hose & this happened went to remove the temp tubing setup & was useing sissors (I know 1st dumb thing it did) to cut off the tubing on the barbs > stabbed the rad "Holey S**T " but it looked as if i missed the tubes just fin damage
SOO I ran house water thru unrestricked & counldn't tell for sure if it was leaking
SOO >>> I think >> I'll plug the outlet & then I'll know for sure . Soo I hook it all up & place a plug on the outlet hose Then I run in the house turn the water back on >> by the time i got 1/2 WAY back i could hear the fins popping ..


>>Everyone Laff now ! >>

>> GTXBlkIce 480 shot to Sh**

it is now > leaking ... but NOT at the site of the damage !


link to the post I did with the pix of damage ....>>>>> http://www.xtremesystems.org/forums/...d.php?t=187491

OH well $140 down the toilet & a valuable lesson learned

NOTE to myself Never agin test anything & I do mean anything on house water pressure !

[Elusid]

That sucks, so when are you ordering another one? lol

[bigslappy]

That sucks, so when are you ordering another one? lol

TOMORROW !

[Pedalmonkey]

the cooling on this thing should be ridiculousness. 1000CFM on a huge high flow designed rad. just might have to do that for my window

[Andreoid]

love this radiatorbox

[[ShowMe!]]

Why not use 4 x 220 CFM deltas on the rad...

That would eliminate all that construction work, and take up alot less space.

And you would have 880 CFM per rad...

Im assuming that sound is not a concern in this build....

[bigslappy]

Why not use 4 x 220 CFM deltas on the rad...

That would eliminate all that construction work, and take up alot less space.

And you would have 880 CFM per rad...

Im assuming that sound is not a concern in this build....

amazmingly low noise comes from this blower a slight hum is all that is heard & the rads made no noise & the fact that i can run it on 110 a/c & not have a PSU just for the fans as all this under the floor to get the cool 50F air

[bigslappy]

got the replacement rad !
back to building this weekend

[bigslappy]

this was posted on another thread
Quote:____________________________________________ __________________________________________________ ________________
Originally Posted by Soulwind
There's another thread about setting up a crawl-space cooling loop on here.

One thing that's bugged me about the idea (in both threads now), is this:

If the crawl-space ambient temp is really that much cooler than the ambient temp in the room the computer is housed in,
and the loop works to bring temps down closer to the crawl-space ambient, then wouldn't you have to worry about
condensation just like the chilled-water/phase-change cooling setups do?????? I don't think I've seen anyone mention that before.
__________________________________________________ __________________________________________________ _______________
Humm!
we'll see when i get to that point but I'm in California & the humidty is not a problem like in other areas but i am addressing that issue with the next stage
as the cooler is a mock up stage right now ( works great) & have now plans of going forward with a sheet metal version BUT 1st i'll do testing on my test bed
__________________

[mike8913]

you shouldnt have to worry about condensation since your change in temperature is so minimal.